Podcasts about ptq

Watch the full episode on YouTube:We first covered Baseten last year when DeepSeek mania was at peak hype. Now they have raised a monster $13B round and become one of the new cohort of AI Infra decacorns that are (with Nvidia, Intel, and the semis complex) chief beneficiaries of the Inference Inflection. We return to Baseten at the peak of the 2026 edition of Open Weights debate. Ali has published a viral breakdown of Kimi K3:And since you last saw him, Philip has spoken at AI Engineer and written the definitive book on Inference Engineering spotted all over SF:Three years ago, inference engineering barely existed as a category.Today, it is one of the most critical disciplines in AI. Inference engineering inherently tackles a different question than standard model training: “How do you turn those weights from training into a product that is fast, reliable, and affordable at scale?” Focusing on these creates an entirely new optimization problem.In one recent GLM-5.2 experiment, quantizing more of the model actually preserved its benchmark quality while increasing throughput by 20%, because the errors introduced in different layers could cancel each other out.Inference is no longer just the final step after training. It is becoming its own engineering discipline, with its own research problems, infrastructure, and increasingly specialized roles.In this episode, Baseten's Philip Kiely and Ali Taha join swyx and Vibhu to explain what actually happens after a new open model is released and what it takes to turn “we generated a token” into a fast, reliable, production-ready API.We go deep on cache-aware routing, disaggregated prefill and decode, quantization, speculative decoding, KV-cache movement, model parallelism, GPU kernels, and the race to make frontier models up to 10× faster. Philip and Ali explain why inference optimizations can still produce gains of 20%, 100%, or even 200%; how quantization errors can cancel one another out; why identical weights can behave differently across clusters; and how Baseten grafted a Kimi vision encoder onto GLM-5.2 without changing the underlying language model.The conversation then expands beyond LLMs into NVIDIA Dynamo, mega kernels, Rubin, AI-specific chips, local inference, video generation, diffusion versus autoregressive models, and the enormous compute barrier to generating coherent long-form video. Finally, we explore the convergence of training and inference, continual learning through persistent KV cache, and the emerging loop where models help optimize the infrastructure that runs them.We discuss:* What happens when a 200,000-token request enters an inference system* Cache-aware routing and reusing previously computed KV cache* Why prefill and decode are increasingly handled by different GPUs* When dedicated deployments become cheaper and more reliable than shared APIs* How speculative decoding uses a smaller model to accelerate a larger one* Tool calling, structured outputs, and what LLMs actually do* What it takes to support a new open model on day zero* Grafting Kimi's vision encoder onto GLM-5.2* Retrofitting inefficient model layers with components from other architectures* Why models sometimes collapse into repeating the same token* How hardware, kernels, and race conditions create nondeterministic failures* Preserving model fidelity while making inference faster* How quantization errors can cancel each other out* Why inference optimizations still deliver gains of 20%, 100%, and 200%* How optimized serving can make a model up to 10× faster* NVIDIA Dynamo, KV-aware routing, and distributed model serving* Speculative decoding the speculative decoder* Why local AI is about making models less dumb while data-center AI is about making them less slow* Tensor, expert, and pipeline parallelism across GPUs* Hardware-aware model design, auto-tuning, and the case against mega kernels* Rubin and why inference is becoming a systems problem* Whether modern GPUs are evolving into programmable AI ASICs* Why enormous models like Kimi K3 require GB300-class hardware* Why open-source video generation still trails Veo, Kling, and other closed models* The quadratic attention bottleneck behind long-form AI video* Autoregressive video, real-time generation, and compounding quality drift* Why future video systems may combine autoregressive and diffusion architectures* Training for inference and inference for training* Continuous post-training, deployment, evaluation, and improvement loops* How GLM-5.2 helped optimize the kernels serving GLM-5.2 itself* Why faster networking could unlock dramatically faster decoding* Continual learning, KV-cache compaction, and persistent model memoryShow Notes* How to build a day-0 API for Kimi K3* 22580: From GPT2 to Kimi3, ExplainedPhilip Kiely* LinkedIn: https://www.linkedin.com/in/philipkiely* X: https://x.com/philipkiely* Inference Engineering: https://www.baseten.co/inference-engineering/Ali Taha* LinkedIn: https://www.linkedin.com/in/aliestaha/* X: https://x.com/waterloointernTimestamps00:00:00 Introduction and the 200K-Token Prompt00:03:18 Dedicated Deployments, Speculative Decoding, and Tool Calling00:11:26 Launching Production-Ready Open Models00:19:06 Model Retrofits, Failure Modes, and Nondeterminism00:28:22 Quantization and Canceling Errors00:32:15 The Race to 10× Faster Inference00:40:48 Dynamo, Speculation, and Local vs. Data-Center AI00:50:18 Model Parallelism, Auto-Tuning, and Mega Kernels01:00:55 Rubin, GPUs vs. ASICs, and Custom AI Chips01:10:03 Giant Models and the Limits of GPU Memory01:12:42 AI Video, Quadratic Attention, and Autoregressive Generation01:21:47 Audio, Images, and Diffusion Models01:27:32 Training, Self-Optimizing Models, and Continual Learning01:40:06 Closing ThoughtsTranscriptIntroduction: Baseten, Waterloo Intern, and Inference EngineeringSwyx [00:00:00]: Okay, we're here in the studio with Philip, old friend from Inference Engineering, the book, as well as Baseten and everything that you've done, you and I have done before, as well as Ali. Welcome.Ali [00:00:15]: Pleasure to meet you.Swyx [00:00:15]: Waterloo intern.Ali [00:00:16]: Waterloo intern, always.Swyx [00:00:17]: When did you get “Waterloo intern” as a handle?Ali [00:00:19]: As a handle? Oh.Ali [00:00:20]: I think the rebranding happened mid-March. When I saw it was open, I was like, “I have to take it. Up for grabs.”Philip [00:00:26]: The problem is that Ali is really good at his job and is not gonna be an intern much longer.Philip [00:00:30]: So we have to figure out who's gonna get the handle.Ali [00:00:33]: Well, I'll pass the torch over to the next intern.Swyx [00:00:34]: Oh, okay. It can be, like, you just pass it to another Waterloo grad.Ali [00:00:37]: To another Waterloo intern. No, bruh.Philip [00:00:39]: Yeah.Ali [00:00:39]: Intern.Swyx [00:00:40]: Intern, yeah.Ali [00:00:40]: And no.Philip [00:00:41]: You gotta get an intern from Waterloo.Ali [00:00:42]: Yeah, I've gotta get an intern from Waterloo.Swyx [00:00:44]: Right.Ali [00:00:44]: But they have to follow the path.Swyx [00:00:45]: Oh, it could, but it could come from Baseten, so it's like whoever Baseten gets from Waterloo.Ali [00:00:48]: Right.Swyx [00:00:49]: Has the title of Waterloo.Ali [00:00:50]: It stays in the ecosystem.Philip [00:00:51]: Exactly.Ali [00:00:52]: Halfway through the internship, you either get it or you're out.Philip [00:00:55]: You should also do, like, a big graduation ceremony where you change the handle.Ali [00:00:59]: Just say it.Philip [00:00:59]: For everybody.Swyx [00:01:00]: You guys are good at ceremonies, clearly. We had a nice launch of the book, very successful. But before we get into all that, I wanna start off with a fun question for you. Okay, you're an expert inference engineer. What happens when I send a long query, say two hundred thousand tokens into Baseten's inference? What's the process of query through GPU model routing, balancing, all that? What is all the stuff that we don't think about?Long Context Requests, KV Cache, and Cache-Aware RoutingPhilip [00:01:26]: With a long query specifically, the first thing that I'm gonna ask is, “Have you sent me this query before, or at least part of it?” and I really hope you have, because it's gonna be a lot easier for me and a lot cheaper for you. So the first thing that we're gonna look at is some cache-aware routing, where we're going to see, we probably have a number of instances, a number of replicas up serving whatever model you're hitting. We want to send this one to something with, number one, available prefill workers, and number two, ideally some cached input already there so that we can skip prefill on at least part of these two hundred thousand tokens. If you're doing two hundred thousand tokens, it's probably coding or a multi-turn agent or something where you would expect to have that cached. If you don't, we're gonna have to send it to a prefill worker. We've at least on certain models disaggregated prefill and decode, so you're going to have one set of GPUs that's solely going to process the input, create the KV cache, and get you your first token, and then that's going to be passed over to a separate set of GPUs, which is going to run decode. We're going to iteratively make those tokens. We're probably going to have some speculator model in front of that. I'm going to assume that you're doing coding, and because of that, our speculator model, which assumes you're doing coding, is gonna have a high draft token acceptance rate. If I'm wrong and you're asking me to summarize every Harry Potter book, it's gonna be slower. And then we stream that output to you and account for it, charge you, a couple of pennies and say, “Hey, would you like to send another one?”Swyx [00:03:04]: Except Baseten doesn't charge by pennies.Philip [00:03:07]: Well, yeah, we charge. I'm assuming that we're talking about the public model APIs. If you are setting up a dedicated deployment, then yeah, it's not pennies.Public APIs vs. Dedicated DeploymentsSwyx [00:03:18]: Yeah, one of the key differentiators when I was talking with Baseten initially was that people who want very high volume just need to rent by the box, ‘cause then it's up to you to figure out how to saturate the box.Ali [00:03:31]: And more often than not, it's, like, way cheaper if you're pushing, like, millions of tokens per hour, if you just pay per hour instead of pay per token.Philip [00:03:37]: Yeah, they do. I think that we've increasingly seen a lot of demand for the pay per token APIs, just because everyone wants to try open models, and then once they find a use case that's really sticky, then they move over to dedicated.Swyx [00:03:51]: Is there a best practice on when it's time to swap over?Philip [00:03:54]: Couple reasons. Yeah, reliability, that's a big one, right?Ali [00:03:57]: Like, if they have a very specific use case, they want you to train something specifically for them, like they want their own spec dec, for instance, for their own traffic.Swyx [00:04:04]: Spec dec is speculative decoding.Speculative Decoding and Custom SpeculatorsAli [00:04:05]: Speculative decoding, yeah.Swyx [00:04:07]: You have to explain.Ali [00:04:07]: Sorry. Like, speculative decoding is like, if you have a huge model, right? And so the model is going to be generating one token at a time every single turn, every single forward pass. So we attach, like, this little, like, parasite, like this layer that goes on top of the model, and this model just has to predict. It does three very fast autoregressive forward passes, and it will predict, like, three certain tokens, and then you do one forward stage over the entire original model in order to see if those predictions were correct or not, and then you accept them or you reject them. Now, this draft model is traffic specific, so if you, like, Philip said, if you're summarizing Harry Potter books, I can train exclusively that draft model on Harry Potter books, and I can guarantee you that I'm gonna accept the three tokens every single time. And so with that case, I increase your decode speed. I wouldn't be able to provide this to you if you're a shared endpointSwyx [00:04:53]: YeahAli [00:04:53]: ‘cause I have no idea if you're doing Harry Potter, if you're doing coding, if you're doing English. We don't know. Also, there was a thing in the book that mentioned that if they really cared about a specific threshold, chapter four, I think. Do you remember that?Philip [00:05:06]: Yeah. The things that you can do is you can set a specific, like, batch sizing, a specific, like, parallelism strategy if you're trying to optimize for, like, throughput versus latency. You can. Maybe a NVFP4 quant doesn't pass your benchmarks and you wanna run a model at higher precision, you could do that. There's just a bunch of reasons why you might wanna have your own endpoint and the biggest one, of course, just being, like, you don't have to deal with someone else doing a hundred million tokens of benchmarking traffic at the endpoint when you happen to be trying to serve your users.Swyx [00:05:40]: Yeah. I think one thing that is. That is a classic journey. Like, it's people is asking the, what happens when you type Google into the browser. Tool calling, is that just, you're generating JSON or is there more complication beyond that?Tool Calling, JSON, and Structured OutputsAli [00:05:58]: Certain customers that we have, they have their own post-trained models, and so they demand a tool calling that's not just, like parse a file or go find the weather. It's something that's very specific and you have to do post-training on this. And if the post-training on the model is not good or if the quantization after the post-training to get the inference to be fast, the model will struggle reading the JSON file and reading the tool calling. But it doesn't require its own like sandbox. It's not like it's going to use that tool calling to like escape a sandbox or like it doesn't have to be contained. It can just be a normal dedicated deployment. The challenge with tool calling more and more seems to be that the companies want certain tool calling which is a very sensitive thing to train. And because you're dealing with all of the JSON outputs, if it doesn't like close the end of the request in a very certain manner, you end up with a model that did the tool calling and like the thinking and so as a result of that, it didn't see the result and just hallucinated the result as it decoded. That seems to be the most challenging thing with tool calling, not really the sandboxes model.Philip [00:06:56]: Yeah, that's a challenge on the training side and then on the inference side, there's work that you can do to scope the possible output. So we published this at this point close to two years ago, the solution to this problem which is you make a state machine and you use that to constrain the output to a specific format. So this is the structured output problem. If you remember backSwyx [00:07:27]: Yeah, the specific grammar is,Philip [00:07:29]: Yeah, exactlySwyx [00:07:30]: GML had this thing.Philip [00:07:31]: Yeah. So it's like the old-school “make sure this is only JSON”, return only JSON orSwyx [00:07:38]: YeahPhilip [00:07:38]: Grandma's gonna die type of prompts.Swyx [00:07:39]: Is it BNF grammar? At some point OpenAI had released a thing that was like, yeah, if you want to constrain your output, write BNF grammar, back as NOR.Philip [00:07:47]: In our inference system, it's just a specified output format. And you get the guarantee that your output's gonna be structured along that format. And so applying that to tool calls can like help cut down on. You can still call the wrong tool or call no tool. It doesn't solve the certainty problem but it at least solves the output structuring problemSwyx [00:08:10]: YeahPhilip [00:08:10]: Within tool calls.Swyx [00:08:12]: And MCP is just another form of tool, right.Philip [00:08:14]: Yeah, exactly.Swyx [00:08:15]: As far as there's no special thing there.Philip [00:08:16]: The thing I'm always like explaining to people is the LLM is not capable of doing anything. It's only capable of making suggestions of what to do and then if those suggestions are formatted in a certain way and applied to a system that knows what to do with them, then an action occurs.Swyx [00:08:32]: Yeah. Part of the fun stuff is, this is solved outside of tool calling too. Like in an agent loop if the output is not correct or you're right, like reasoning, tool calling was done in the reasoning trace, just be like, “Oh, I don't know what to do. Let me just try again.” And it might get there after a few tries. And on your point of training, sometimes this is harder in smaller models, so you don't have the same exact quality outputAli [00:08:56]: Right.Swyx [00:08:57]: When you just swap from a big model, right?Ali [00:08:59]: Yeah. I will say that, before, I think we need to go back to inference engineering proper.Ali [00:09:04]: But, I had expected that something would replace JSON because it's hard to stream JSON ‘cause JSON must be complete and you must have open and close brackets and everything. So it's hard to parse something or validate something while it's being streamed. So people invented all sorts of things that are like, I forget the name of some of these alternatives, but it's something like TOML, something like YAML. But JSON seems to be dominant still.Philip [00:09:30]: The JSON outputs aren't that long, right? Like you could have a long-- ‘cause tool calls also contain the arguments in them and perhaps for a certain tool you might pass like a very long argument. But my impression of the median tool call is that it's a relatively small number of tokens, right? So I would expect that speculators are generally fairly good at something as formatted as JSON. And so you would have like a pretty fast decode step there and that the streaming wouldn't be as valuable, but maybe I'm wrong about that.Ali [00:10:02]: I think you're also bounded by the software or that the model is gonna integrate with if the software is built with JSON for the tool calls or if the company that you'- if your customer says that this is how our software works and our tools are interfaced with JSON, you can ask them to like, change their software and say like, “Yeah, this is gonna be better for the model.” but like with the right training shouldn't be that much of a difference. Also more profitable if it outputs more tokens probably.Swyx [00:10:25]: Depends on your business model.Swyx [00:10:27]: It really depends. But I will say that, as a writer with like experience a lot with generated output, I do try to move from text to JSON text which is very long JSON, right? Like there's paragraphs in every field because I'm trying to structure it, right?Philip [00:10:44]: Right.Swyx [00:10:44]: I want you to first make factual statements, then make opinions then make bullet point summaries, have dates, have entity references have your sources for references, all these things. Anyway, so these are things that like I think people who really experiment with structural output have to really care about. But, let's, let's recurse up the stack a little bit. Before we started recording, you mentioned something really cool, which is that there's a lot of engineering that-- inference engineering that goes on when a new model provider releases a new model, right? So let's call it GLM-5.2, Kimi K3. I had previously assumed, especially if it's like, well, GLM 5 to 5.1 to GLM-5.2, like that you've supported them before. Is it that much work?What It Takes to Support a New Open ModelAli [00:11:26]: It's a lot of work.Swyx [00:11:28]: Yeah. Okay. So like, a lot of people, all you guys, right whenever a new model launch like, people rush to say like, “Oh, Hugging Face supports this, Fireworks supports this, Spacetime supports this,” and I'm like, “Yeah, of course we support it.” But what goes into that? What goes intoPhilip [00:11:40]: I think it's more than just support it too, right? It benefits the consumer a lot. Like I think it was with Kimi K2.5 or GLM-5.2 the latest, there was an inference war, right? X provider is at 90 tokens a second. The next day we're at 150. The nextSwyx [00:11:55]: I kinda kicked that off with the GLM-5.2.Swyx [00:11:58]: I wrote a Twitter article about. It got like half a million views,Ali [00:12:02]: Based on being numberSwyx [00:12:03]: YeahAli [00:12:04]: Or it's for something else.Swyx [00:12:05]: Yeah. Which,Ali [00:12:06]: Oh my GodSwyx [00:12:07]: Which then got everyone really excited about, hey, how can we, bend tracks a little bit further and,Philip [00:12:14]: There's a difference between support the model, as in I can make a token out of this model, and support a model, as in I have a production-ready API from this model.Philip [00:12:26]: Getting to the point of I can make a token out of this model is not that hard because generally the, open source inference engines, vLLM, SGLang of the world oftentimes even receive weights ahead of time, maintainers do, or the people making the model merge PRs to ensure support. So you generally can, just get it working on the standard open source stack without too much pain in most cases. The challenge is, every inference company is gonna have own proprietary stack. Some open source components, some in-house stuff. And for any arbitrary model, there's going to be some new stuff. Sometimes you get lucky, like K, two five to two six was, like, pretty similar.Quantization, Speculators, and Production ReadinessAli [00:13:16]: Yeah. It was pure continued post-trainingPhilip [00:13:18]: YeahAli [00:13:18]: If I remember correctly.Philip [00:13:19]: Even in those cases, there's still stuff you have to do. You have to redo the quantization work. You're taking the model from. Generally, these models are not released in NVFP4, and we want them to be in NVFP4 for maximum Blackwell compatibility. So we have to perform that quantization, and, calibrate the quantization to make sure that we're not causing any regression in the model's intelligence. And then we also have to train the speculator, as we've talked about. Generally, we have. We have ZDR, zero data retention on our model APIs, so we don't know exactly the traffic that people are sending us, but we know what's popular. We know that coding use cases are popular. We know that agents, agentic use cases are popular. So we can get public data sets that are representative of that traffic and train general speculators. Now, with speculators today, you need to train the speculator using the base model itself because you're getting hidden states out of the model from running inference on these specific prompts, and that is the training data you use to create the speculator. So there's that process which you need the real model weights for. And then there's of course just the process of, standing up all the infrastructure behind it, loading all this stuff, testing it. And then when there's a new model with a newer architecture, I think that, like, the DeepSeek models tend to be the most challenging as they have, like, the most novel architectural stuff going on, model after model. But every new model has something. Kimi K2 had. Oh, sorry, GLM-5.2 hadAli [00:14:53]: Sparse attention.Philip [00:14:54]: Yeah,Ali [00:14:54]: YeahPhilip [00:14:54]: the DSA.Ali [00:14:55]: Right. Which is brought from DeepSeek.Philip [00:14:57]: Yeah. AndAli [00:14:59]: So you can copy-paste then?Philip [00:15:01]: It kindAli [00:15:01]: I don't know how this works.Philip [00:15:02]: So, like we had to, like, build support for that into our runtime. And you're right, like it is really interesting the way that all of these open source labs borrow from each other. For example, like GLM-5.2 doesn't have vision. So something that, Haley, a guy on our team, if we could take a look at this, he, like, grafted the Kimi vision encoder onto GLM-5.2.Retrofitting Vision into GLM-5.2Ali [00:15:27]: We'll be training the projector.Philip [00:15:28]: Exactly. So if you think about, like, the encoder, there's the encoder, which is the part that looks at the image and turns it into latent information, and then there's the projector which likeAli [00:15:38]: You can say latent space. It's okay.Philip [00:15:41]: And then there's the projector that maps it onto, the model itself, and then there's the model weights. You don't wanna mess with the model weights because you run a chance of making the model dumber at something else for the purpose of giving it vision. So instead, Haley started with just a projector, which is only a handful of millions of parameters.Ali [00:16:02]: That would be, yeah.Philip [00:16:02]: Yeah.Ali [00:16:03]: Can you show the training one?Ali [00:16:04]: Like the way it groksPhilip [00:16:05]: YeahAli [00:16:06]: Very interesting.Philip [00:16:06]: And maybeAli [00:16:07]: That right therePhilip [00:16:07]: Maybe Ali, you should take it from here. You've got a betterAli [00:16:10]: Ooh, double the sandPhilip [00:16:11]: Understanding of this than I do.Ali [00:16:11]: Yeah. You can see, like, he. The way he trained this is really cool. At the beginning, he was training it using just like, “Here's a picture of a mountain. Can you describe what's in this mountain?” And that caused it just like the first, learning walls. Like here you can see this all we're trying to teach it is to translate the encoded. Like it's already taken the encoder from Kimi K. It's taken the image. It'Philip [00:16:31]: Yeah. FrozenAli [00:16:31]: FrozenPhilip [00:16:32]: With adapter.Ali [00:16:32]: Exactly.Philip [00:16:33]: Yeah.Ali [00:16:33]: So the brain is frozen and the eyes are frozen. It's just we're tryingPhilip [00:16:37]: AlignAli [00:16:38]: Interconnect between the eye and the brain, right? So the projector. And so you take the tokens and then he's like, “Oh, can you describe what's in this image?” And he's like, “Oh, it's a mountain,” or it's a person or it's a human, whatever the case is. But that didn't cause complete understanding. So he changed it such that every image was associated with a data set of questions. Like, does this image have a white male? Does this image have birds in the top corner? Does this image have a scientist in it? All of that stuff. And it would have to answer questions correctly. And using not just training on describing an image, but being able to answer question, another question, answer over time. Like you can see the grokking, which is like genuinely insane, that retrofitting vision into a large LLM can learn to that extent. And even for images that it doesn't perform well on, for instance, if you ask it a picture of like Stephen Hawking, “Who is this?” Maybe it doesn't get it, but it will say something like, “This is Albert Einstein.” Like it still understandsPhilip [00:17:25]: Close enoughAli [00:17:26]: That this is a scientist who is a man who has, some significant achievements, all that stuff. So that's like really cool.Philip [00:17:32]: Yeah. So, we've covered Hao Tian before, who the author of the LLaVA paper that did this, a while ago. And I think that's very foundational work for anyone who hasn't done vision work before.Ali [00:17:41]: Same with the CLIP and MetaCLIP, where you go from just captioning to building out questionsPhilip [00:17:47]: RightAli [00:17:47]: Off the image and how much better you can get performance.Philip [00:17:50]: Right. Right. Right. Yeah. But what's, what's so exciting about this is if you look at a model like this. Now, this is a little bit more of a research project. It's not. It got to 56% on MMLU Pro, I think. So not quite frontier. But if you're running this model, you haven't suffered any loss on your GLM-5.2 quality. If you don't have an image, it'll just behave exactly the way it used to. And ultimatelyAli [00:18:14]: Which in the inference code you literally do not include the other part, right?Philip [00:18:18]: Yeah. You would just skip the encoder if you don't have an image input.Ali [00:18:22]: Okay.Philip [00:18:22]: Just confirming.Philip [00:18:23]: YeahAli [00:18:23]: Does it affect a lot on the overall inference side? Like you're not adding much, you're adding a very small vision encoder. These are typically likePhilip [00:18:30]: They're super fineAli [00:18:31]: Less than a billion parameters, right?Philip [00:18:32]: Yeah. It's, - There's a little bit less standardization among vision encodersSwyx [00:18:37]: YeahPhilip [00:18:37]: So the support matrix can be a little bit, sparser. But overall, yeah, it's a pretty, it's a pretty minor component of the overall system. And ultimately what you get out of the system is all of a sudden you have Kimi Vision, GLM weights, and DeepSeek attention all in one model.Open Source Model Grafting and Franken-MergesPhilip [00:18:56]: And that's, I think, a lot of the power and beauty of open source, is that you can take all of these different components and combine them together into a system that's better than anyoneSwyx [00:19:05]: YeahPhilip [00:19:05]: Can be individually.Swyx [00:19:06]: People used to say that you would also do Franken-merges where you would take likePhilip [00:19:10]: YeahSwyx [00:19:10]: Layers from each model.Swyx [00:19:11]: Does anyone do that anymore?Ali [00:19:13]: Well, to your point previously when you were mentioning like, the work that goes into supporting a model when it first comes out, like GLM-5.2 or MiniMax M3 or whatever the case is. Sometimes you do have to like, you do have to switch out some things. Like, for instance, the MiniMax M3 head uses full attention, and with full attention you end up with this like insane bottleneck in spec dec ‘cause you're doing auto-regressive token generation for three tokens, and you're doing this like N squared over all of the tokens that are in your sequence. Your KV cache is like very large because it's not sparse, it's not top K. So we find it better to like, okay, we're gonna replace this, we're gonna replace this layer with a layer from another model that's using like GQA, for instance. And then just with the right training, you can get it to have the same acceptance rate. So it is very possible to retrofit layers from other models and very much needed. If a layer is like inefficient, the training just becomes the challenge, like how do you ensure that you train it properly? Which again to your earlier point is like the mesh between training and inference. As in like you need very good training in order to do fast inference. That's like, I feel like more and more becoming true.Swyx [00:20:21]: Yeah. Anything else on the support side when you say like get it to fully production ready?Loop Detection, Race Conditions, and Non-DeterminismPhilip [00:20:26]: Yeah. I think that there's also a question of just, we can test a model to a pretty extensive degree, but we're trying to get it out quickly and then you see a bunch of other people test it and you get interesting results. There was an issue with, GLM briefly where we had some like mode collapses where it would just output the same token over and over again for certain prompts on certain temperatures. Like once you expose an endpoint to the real world, there's going to be, so many more varieties of things given to it that you're able to, discover and patch things. So it's not just a, day zero process, it's then like for the first week, for the first month, if a model remains popular, like how do you both fix bugs and then continue to push the envelope on performance?Ali [00:21:21]: What do you mean you don't want your model outputting S?Swyx [00:21:24]: Is there loop detection on that stuff, by the way? It still happens like quite a lot, which is surprising.Ali [00:21:30]: We have like we, in our endpoint, like if a model was to output the same token like four plus times, we just cut the generation. We say like, “Oh, sorry, this-- Like try again,” or like we will reprocess the request. ‘Cause we know then, like if it, like if, yeah, it's four times the same token, it's probably collapsed.Swyx [00:21:45]: Yeah. Is there a way to opt out in case I really want that?Ali [00:21:48]: You want that?Ali [00:21:50]: I think there's a way that we have to handle it. I'm not exactly certain, but I feel like in certain models, like when they output something like you can imagine, like a table for instance, and so they want, they wanna draw like 12 dashes and 12 dashes. Yeah, I think there's a way for that to happen. I think we only do it on certain tokens. Like we exclude certain special characters.Swyx [00:22:07]: Yeah.Ali [00:22:07]: So we only do it on like certain like S is the most common almost. GLM-5.2Swyx [00:22:11]: OhAli [00:22:11]: And I think it was DSV 4 as well. Like you'd just have like looping issues where like you literallySwyx [00:22:17]: ItAli [00:22:17]: Just have like S.Swyx [00:22:18]: Yeah. Is there a special, something special about S? No, just randomlyAli [00:22:21]: It just seems to be the one token involved.Swyx [00:22:23]: Yeah. And it'Philip [00:22:24]: Is thereSwyx [00:22:24]: And it's only temperature 0Ali [00:22:27]: NoSwyx [00:22:27]: Even at other temperaturesAli [00:22:27]: Even at like 0.9 or whatever, it will still, it will still collapse.Swyx [00:22:30]: That's weird, right?Ali [00:22:30]: It's, it is an inference problem to be honest, like a software problem. Like oftentimes, the image you run will-- like NVIDIA will release an image for instance, and if we will upstream the changes from their latest TensorRT-LLM image into our stack, we'll find that it fixes it. Or oftentimes this will only happen in an inference engine that you're using like SGLang. But if you were to switch to vLLM, that isn't the case. So it seems to be like an extremely like deterministic software issue and not really a model issue. It's not like a weights problem. Like I'- we'll say like, “Oh, it's a problem with the quant. We did PTQ wrong,” right? But that isn't, that doesn't make sense because the same weights used with a different inference engine does not repeat the problem. And sometimes it's, the kernels that are being used in the backend have like these very subtle sometimes race conditions, where if you were to use this model hosted on one cluster, you will never get this problem.Swyx [00:23:19]: Oh my God.Ali [00:23:19]: But if you host it on a different cluster, you will. And the reason is the KV cache transfer from a node to node in that one cluster is using a slower interconnect than the node to node in another cluster. So that exposes the race, whereas in another cluster it doesn't. So then you end up just like, okay, this model is not gonna be hosted on this cluster. We're gonna host it on, another cluster because that cluster exposed that problem. But then it ends up with like, okay, is it the software? Is it the model weights or is it the hardware?Swyx [00:23:42]: There is a thing about this with temperature 0 still not being deterministic, right?Ali [00:23:46]: Right.Swyx [00:23:46]: Mostly because of hardware. Even at temperature 0 same model, you won't always get the same output.Swyx [00:23:52]: Even-- But I'm surprised by the race condition one because, I thought PyTorch was a graph that like guarantees that you at least, execute things in the right order.Ali [00:24:02]: Well, yeah, true. Like I'm not, I'm not saying that there is. Like well, you have things like PTL optimizations where like you can start a kernel before the end of the previous kernel, and that's like ‘cause you want to do that because there'sSwyx [00:24:12]: It's like pipeliningAli [00:24:12]: Expense. Exactly.Swyx [00:24:13]: Yeah.Ali [00:24:13]: But it'- But you don't do it cleanly. Like you overlap a little bit of the execution. No, it is very possible that the kernel itself, like that one block that is supposed to be running in this instance of time, that kernel itself has a race condition. For instance, like a missing barrier. Like often if you're designing a kernel and you want it to make it to be very fast, if you don't test it extensively, you'll, you'll have certain threads access data points from registers before they've been written to by other threadsSwyx [00:24:36]: YeahAli [00:24:36]: For example, because like your barrier is wrong or your synchronization was wrong. But yeah, like the testing itself is very difficult in those like, andSwyx [00:24:42]: And there's no like borrow checkerAli [00:24:45]: What does that mean?Swyx [00:24:46]: Like Rust. Like the. If you're trying to have like memory safety It sounds like a comparable problem.Ali [00:24:52]: Well, yes, but you're working in CUDA, right, NVIDIA GPUs. Like- You just need a higher level language like modular Maybe that's what modular is supposed to do. I don't know.Quantization Quality and Vendor FidelityVibhu [00:25:00]: How do you see keeping quality of the model? So you talked about all these steps of, okay, you gotta do quantization, train your own speculative decoderAli [00:25:07]: RightVibhu [00:25:07]: Run on different hardware. Looking at other model providers, okay, you kicked off a inference speed race on the consumer end. What goes into keeping quality the same across them, right? Sure, you can run benchmarksAli [00:25:22]: YeahVibhu [00:25:22]: But, like, how do you determine how much quantization are there standards? What goes intoPhilip [00:25:27]: There's a few things on quality. Most inference optimizations are lossless. KV caching, for example. You are just recomputing or preventing recomputing the same values. Speculation, of course, if a draft token is wrong, it gets rejected. The main lossy optimization is quantization. And that really comes down to, number one, data format, number two, which parts of the model you choose to quantize, which layers, and number three, like doing a lot of calibration on the quantized weights, to ensure that you're preserving all the outliers. There's other tricks that you can do, though. A big one is long context, ‘cause one thing you asked at, right at the beginning is, “Oh, what's gonna happen if I send a 200,000 token request in?” So with a long input sequence, you need to, store a lot more information. You need to process a lot more tokens. And so even if a model has a context of a certain length, you might, as an inference provider, choose to build an API with a shorter context length, and of course a full length one as well. Because if someone doesn't need the full million token context, for example, you can get them better performance. I don't know if that's exactly like quality of the model. The way that I think about quality is to what degree are we faithfully serving the original model? If you think of a golden implementation of a model that performs exactly the way the model is designed to perform, I think of quality as how close are we getting to that, 100% fidelity of the model.Philip [00:27:13]: You can also, of course, think about quality from the training side and how do you push yourself past 100%. But when I think about purely inference optimizations, it's getting faster while staying as close to that 100% fidelity mark as possible. And certainly our standard internally is that, like you should not be able to tell the difference between our API and a, official API. I think Kimi in particular does a good job of vendor benchmarking hereAli [00:27:41]: YesPhilip [00:27:41]: Where they haveAli [00:27:42]: They released an actual vendor benchmark.Philip [00:27:43]: Exactly, yeah.Ali [00:27:44]: ‘Cause they accused, some people, Amazon? There was some provider that was not doing very well on Kimi's benchmark.Philip [00:27:50]: Yeah.Philip [00:27:51]: So, with Reflect we probablyVibhu [00:27:52]: This was a long time ago, right?Philip [00:27:54]: No.Ali [00:27:54]: Yeah, like threeVibhu [00:27:55]: They alsoAli [00:27:55]: Four, five months agoVibhu [00:27:57]: This also happened with, I don't remember which model, but they pulled out quite a few, and then they started a whole chart about this. It might have beenPhilip [00:28:03]: Kimi Vendor Verifier.Ali [00:28:04]: Yeah.Philip [00:28:05]: Yeah.Ali [00:28:05]: Yeah, ‘cause you, ‘cause you'd be pissed, right? Like if you'Philip [00:28:07]: Yeah.Ali [00:28:07]: If like if I'm a consumer and I'm using like Amazon's endpoint for instance, and I've used Kimi and I'm like, “Oh my God, like this is bad,” I'm not gonna say, “Oh, Amazon quantized the model in a bad way.” I'm gonna say, “Oh, Kimi sucks.” Right?Philip [00:28:17]: Yeah.Ali [00:28:17]: So it seems like that makes sense.Philip [00:28:19]: Yeah, they care. They care.Vibhu [00:28:21]: Justifiably.Ali [00:28:21]: Yeah, justifiably.Vibhu [00:28:22]: This is probably a stupid question, but just checking, has anything improved from main quantization?Philip [00:28:28]: Yeah.Vibhu [00:28:28]: Like, is quantization always strictly worse?Ali [00:28:30]: Well technicallyVibhu [00:28:32]: NoAli [00:28:32]: It's a lossy. QuantizationPhilip [00:28:33]: YeahAli [00:28:33]: Is a lossy, it's a lossy implementation.Philip [00:28:36]: Speed improvesVibhu [00:28:36]: Speed improves.Ali [00:28:37]: It the number, likeVibhu [00:28:38]: No, I' always look for inverse scaling laws.Philip [00:28:40]: Yeah.Ali [00:28:40]: Yeah.Vibhu [00:28:40]: This is something I learned from Noam Brown, where like things that normally act in one direction sometimes do.Philip [00:28:45]: Well, technically when you run a benchmark, because these models are deterministic, sometimes your,Ali [00:28:52]: YeahPhilip [00:28:52]: NVFP4 quant is like, two basis points higher than yourAli [00:28:56]: No, it's noise. It's noise.Philip [00:28:57]: Yeah, exactly. I'm like, yeah, it's, it's within. That's why I always say within margin of error.Philip [00:29:01]: And I stopped saying that because everyone assumes that what is, well, within some margin of error, we're barely inside of that to the worst, so we're saying. But yeah, sometimes it's just like, gives you a higher output score. But like Ali said, that's noise. To my knowledge, you're not necessarily making the results better. You're just trying to, again, like keep your fidelity as close to 100% to the original model.Layer Selection, KL Divergence, and Better QuantizationAli [00:29:27]: There is, to your point, research that we did on MP. I don't know if you are able to pullPhilip [00:29:31]: YeahAli [00:29:32]: A tweet we did. One of our research interns, Joshua, I think it's a tweet on how we have 20% better quantized GLM-5.2 than NVIDIA. Essentially what we found throughout like this month research is, okay, quantization is a lossy. It's. You're compressing the data from, occupying 16 bits to occupying, four bits, for instance. And so you're losing some information, and you're trying to minimize that. And so when I say that I'm gonna quantize the model, my job becomes how do I find the layers that I can quantize, and how to find the layers to not. For instance, with image models, I don't quantize modulation layers, and I don't quantize out projections because those two are. Like out projection is what you see as the user. Modulation is what the model sees or understands. Right, exactly. And so to his paper, do you have the. It doesn't have the. Yeah. It's a long paper. I don't know if I can findVibhu [00:30:25]: If there's a part to search or it's probably in the thread.Ali [00:30:28]: It's probably in the thread.Vibhu [00:30:29]: Yeah.Ali [00:30:29]: But the long and the short is it is very possible that quantizing more of the model makes the results. Like if I have a model that I quantize layers one, five, and 10, and another model where I only quantize layers one and It is possible that the model in which I quantized more information is going to perform better because the quantization errors have canceled out. And so what Joshua showed in his mathematical proof where he had like a verifier in, is that you can predict which layers are going to have quantization errors that will cancel out with each other, and you choose to quantize those layers. And so the result of doing this mathematical quantization is you end up with a model that's 20% more quantized than another provider, so you get 20% more throughput of it because there's more layers than running an NVFP4, and your quality is better than that other quant because the layers that you chose to quantize have their errors cancel out, like one layer skewed to the right one layer skewed to the left, one layer skewed to the right. Your final logits distribution is more similar to the original distribution of the model, so you have better fidelity. And so the way we proved this was with KL divergence. So instead of just scoring on the benchmarks, we scored the KL divergence between the logit distribution of the quantized model and the logit distribution of the original full precision model, and we showed that with this technique we get. If your probability distribution on the logits which token it wants to select is more of the same as the original model, you're probably gonna end up staying true to the original model. So yeah, so it seems like previously before this, it seemed like the industry was, well, the more you quantize, the worse it's gonna be, ‘cause the more loss you introduce. That's not exactly, not necessarily true. So yeah, doesn't improve it, but can cancel out.Philip [00:31:57]: I think it might be this, but reminds me a good bit about pruning where you can prune off certain layers.Philip [00:32:03]: But very interesting. Didn't know this was a whole paper you guys put out.Ali [00:32:06]: It's. Fun fact, it was originally 72 pages, this paper, and then we decidedPhilip [00:32:11]: WowAli [00:32:11]: We can't tell. We couldn't release it. So it's now 45.Swyx [00:32:15]: Still 39 pages, so very substantive. We talked about evals and all these things and, like what's possible in terms of speedup? Like it's like probably like the numberInference Speedups and BenchmarkingSwyx [00:32:25]: Thing that people do wanna care about, and it's something that you wrote about in your post. Like official API is 70 tokens per second, and you push it up to 90. Is that like a normal thing?Philip [00:32:36]: So what's cool about working in inference, the reason that I think inference is going to be a useful place to do engineering for a long time, is that if you look at highly optimized domains like, say, finance, if you're in finance, you measure how much better you got in basis points. It's like, “Oh, I got five basis points better, like twentieth of 1% better,” that's huge news because everything is so optimized. When we publish optimizations, it's 20%, it's 100% it's 200%. So there's still probably like a lot further to go, honestly. Like you'll, you'll know that inference is pretty much solved when researchers start publishing about how they got 1% faster at something.Swyx [00:33:19]: Which by the way, because I am from the finance background, in the ‘70s, that was the margin at the time. When you did quantitative finance research, you would findAli [00:33:27]: And like 20%, tens of percent.Swyx [00:33:29]: That's. Yes.Philip [00:33:29]: Yeah.Swyx [00:33:30]: And now it'Philip [00:33:31]: Tiny fractionsSwyx [00:33:32]: For those people interested, look up Andrew Lo's paper. He had a really interesting illustration of quant, stat arb, distribution, narrowing down from like those kinds of 20% differences in the ‘70s, down to nothing today, which is very cool.Philip [00:33:48]: Exactly, and we're at the beginning of the same type of thing. Now benchmarking is hard. I think anyone will tell you that, and benchmarking provider speeds is hard because there's so many variables that go into it. What hardware are you using? How much load do you have on the system? What's the exact nature of the prompts and input and output sequence lengths? All that stuff. But overall, when you start stacking these improvements, you're looking at multiples. You can look at it. The most common form, of course, is TPS, tokens per second, which is bad naming by us in the industry, ‘cause there's two tokens per second. There's tokens per second, the throughput number, and the latency number.Ali [00:34:31]: TTMT, yeah.Philip [00:34:32]: Like total tokens per second out of the, out of the GPU as a throughput number. Most people only care about tokens per second as the latency number, which we should call ITL, intertoken latency, but we don't.Philip [00:34:44]: Anyway, so you can imagine a standard API without many optimizations for a 1 trillion parameter model operating somewhere in the 30 to 50 tokens per second range for reasonable traffic profile. And we generally see the goal of, pushing to 10X that. But, not necessarily day zero, but by stacking enough optimizations, if you have, say like four optimizations, each of which doubles performance. Or sorry, three optimizations, each of which doubles performance, then you stack that up, that's an 8X gain. That's the order of magnitude that we're working with in this space. We're trying to make things substantially faster, not just go from like 70 to 90.Swyx [00:35:38]: Are you saying you've. You have done that?Philip [00:35:40]: So let's say you have as a reasonable baseline, 30 or 40 tokens per second. You can achieve 10X that. So like on GLM-5.2, if you run it unquantized, perhaps on H100s even, and you're just using an off-the-shelf inference engine with no particular optimizations, no speculator, nothing extra around like KV routing, no disaggregation, you're, you're probably, yeah, looking at that like 30 to 40. You think that's like a reasonable baseline?Swyx [00:36:12]: Right. Right.Philip [00:36:12]: To get to something like 10X, there's a lot of trade-offs that you're making. If we're running at more like a 300, 400 tokens per second range, you are using the best hardware possible. You have a optimized speculator. You have done all of your quantization work. You are Seeing a pretty high cache hit rate. You are running with a reasonably small batch size and a parallelism configuration that is tuned for latency versus throughput, but it is possible. So the spreads that you see if you, like, go on artificial analysis or you go on OpenRouter and you look at, the worst provider to the best provider, oftentimes can hit that range. 10X is of course very aggressive. It's oftentimes maybe more of a four to six times improvement. But that's the performance that makes us really excited, is when we can get these huge gains, not just go from 70 to 90 tokens.Stacking Optimizations: NVFP4, Speculation, and DisaggregationAli [00:37:19]: It's also, like, hardware dependent. Like, ifPhilip [00:37:20]: YeahAli [00:37:20]: If you have a thing where you're serving it on just, like, a node of H100s and then you throw, like, you shard the model across, like, four nodes of B200s. Like, you can definitely increase the speed with just throwing more hardware at it. Like, normalizing for the same exact hardware and the same number of GPUs.Philip [00:37:35]: Yeah. Then you're looking at, like, a two to 4X improvementAli [00:37:38]: Right. RightPhilip [00:37:38]: Depending on the inference optimizations. So yeah, it's. Some of it's, what's the call, and some of it's who's the driver.Vibhu [00:37:46]: If you break down the two to 4X, say the example is run GLM-5.2Ali [00:37:51]: YeahVibhu [00:37:51]: On B200sAli [00:37:53]: YeahVibhu [00:37:53]: Single node, right? What's, like, the cost trade-off for effort to get, like, the last bit of juice out versus what should people just think of, right?Ali [00:38:01]: Spectre quantization. Yeah.Vibhu [00:38:03]: Spectre quantization.Ali [00:38:04]: That's, that's, that's like 95%. LikeVibhu [00:38:06]: And how far does that get you? And how easy is that for the average person to do? So say right I wanna throw the weights of GLM-5.2 on a node of B200s, how easy is it to find speculative decoder- decoder model or already quantized model? How much work goes into it?Philip [00:38:23]: If you're doing it up front, it's quite a lot of work. If you're doing it today, there's going to be people who have published things that you can just, you can just grab some NVFP4 weights. You can grab a speculator. Yeah, if we're thinking about, like, what are the 2Xs we're stacking, going from, BF16 to NVFP4 is, it's not quite a 2X, right? It's like. I think it's about, like, 30 to 40%, from 16 to 8, and then another 30 to 40% multiplied from, 8 to 4. So that doesn't quite get you a 2X, but, like, roughly a 2X. Speculator, roughly a 2X. Disagg on top of that if you're able to get enough hardware and put enough traffic through it, another roughly a 2X. And then you add in some, double-digit percent increase from having just a better runtime with, the latest kernels and stuff behind it. And that's how it stacks up.Ali [00:39:21]: YeahPhilip [00:39:21]: So building each of those, like, building the, quantized weights is, for someone who really knows what they're doing, hours to days of work. Building the speculator, again, like, hours to days of work. And the, disagg setup, hours to days. Well okay, but like once you haveAli [00:39:39]: Once set up. Once set up. YeahPhilip [00:39:40]: Yeah, getting disagg working for the first time, I'm saying, of course, is very difficult.Philip [00:39:44]: The marginal implementationAli [00:39:48]: Like, if you're just grabbing, like if you are a person, like just a normal consumer who has access to, like, a node of B200s and you're wondering, “How can I just host it myself?” You don't need to quantize the model yourself. There's always gonna be, like, an open source quantized checkpoint. NVIDIA's gonna push one out if no one else does. You. Usually, the providers will have their own spec dec that they've trained as well. You don't need to train your own spec dec. You can just use that as well.Philip [00:40:09]: Yeah. Like, GLM-5.2 has its own MTP.Ali [00:40:13]: Right. Right.Vibhu [00:40:14]: What's multi token prediction?Philip [00:40:15]: Yes.Ali [00:40:16]: I'm justVibhu [00:40:16]: Can you explain that?Ali [00:40:16]: I'm just an expert.Ali [00:40:18]: I can do it for you in case I get it wrong?Vibhu [00:40:20]: No.Vibhu [00:40:21]: Yeah, you should correct if we're wrong, but their multi-token prediction can be used for self-speculative decoding.Ali [00:40:27]: I'm not sure. I'm not gonna correct that.Vibhu [00:40:28]: Okay. I'm semi-confident in thatAli [00:40:30]: Okay. YeahVibhu [00:40:30]: But someone can check. But it's useful to paint the story of, okay, not just the average person, but say a company wants to switch from serverless inference I wanna throw this up on. I wanna rent some GPUs, throw it up. These are the steps you take to do significantly faster than just put it behind vLLM.Ali [00:40:48]: Right.Vibhu [00:40:49]: I was waiting for a mention of Dynamo.Vibhu [00:40:51]: I feel like, that's supposed to be the baseline that you measure against.Dynamo, KV Routing, and Disaggregation ToolkitsPhilip [00:40:55]: I would think of Dynamo as less of a box system and more of a toolkit for building with. So when we talk about doing aware routing, when we talk about doing KV offloading, when we talk about doing, PD disaggregation, Dynamo fundamentally is. By the way, Dynamo is an open source library from NVIDIA.Ali [00:41:17]: We've done a pod with KylePhilip [00:41:18]: OkayAli [00:41:19]: Kyle Cranin.Philip [00:41:19]: Cool. So then your listeners know then that it supports all the different inference frameworks. And it is multi hardware, which is interesting.Ali [00:41:28]: But it's just a router, it's not like an optimizer layer.Philip [00:41:30]: Yeah. All it does, like, what Dynamo is good at, it is a library for moving information around your cluster, around your hardware. So if you have, KV cache on one place and you need it to be somewhere else, Dynamo coordinates NIXL for you to move that around.Philip [00:41:49]: That doesn't mean that, like, out of the box, you just say, “Pip install Dynamo,” and then you get, like, a massive performance speed up. It's more of a developer toolkit.Ali [00:42:01]: Yeah. I would have said it would. It comes with a set of defaults that you can then swap out.Philip [00:42:06]: It does. If the industry at large, I think, was, like, rolling out all of these deployments, standard, then I think it would be, like, a credible baseline. But, we've got to, we've got to benchmark against, like, what we're seeing in the wild.Speculative Decoding Methods: Medusa, EAGLE, n-Gram, and Spec-SpecVibhu [00:42:23]: I did wanna talk a little bit more about PD disagg, because that is probably, like, number three after quantized and speculative decoding. In your book though, I was just gonna pull out the book.Philip [00:42:31]: Yeah.Vibhu [00:42:32]: Like section 522 on Medusa, 523 on EAGLEPhilip [00:42:35]: YeahVibhu [00:42:36]: 524 on gram.Philip [00:42:37]: It's 55, would be disaggregationAli [00:42:42]: Yeah. Well, no, I just wanted to dwell a little bitPhilip [00:42:44]: YeahAli [00:42:44]: The other. Like, so what do you choose to include? What do you choose to not to include? Because there was all these other techniques.Philip [00:42:51]: Yeah.Ali [00:42:51]: Are these still relevant? Because I think they came out, like, a year and a half ago maybe.Vibhu [00:42:55]: Medusa is quite old.Philip [00:42:56]: Yeah, Medusa's old.Ali [00:42:58]: It was old.Vibhu [00:42:58]: But is it in the book as a good, here'sPhilip [00:43:01]: BaselineVibhu [00:43:01]: Baseline vanilla understand it?Philip [00:43:02]: Like you should know this.Vibhu [00:43:03]: Like I read the paper, I'm like, “ it makes so much sense.”Philip [00:43:05]: Yeah.Philip [00:43:05]: So with the book, I had a couple goals. One was to give people just a working vocabulary for the space as a whole, and the other was to give them some intuition about how each of these techniques works. As I mentioned in my AI Engineer talk, which is the first public addendum to this, the speculation space has moved much faster than everything else. So yeah, even at the time that I wrote the book Medusa, I very much included as a way for people to understand how the space evolved rather than what the most modern technique is. And now of course, there's DFlash, dSpark. There's, there's newer techniques even than EAGLE, although EAGLE is still very commonly used.Ali [00:43:51]: SpecSpecta.Philip [00:43:52]: Yes. Speculative decoding.Vibhu [00:43:54]: What canAli [00:43:56]: Oh, it's a paper by Tri Dao and it's like, it's doing speculative decodingVibhu [00:44:00]: HuhAli [00:44:01]: For the speculative decoder.Philip [00:44:02]: Oh, in spec- oh my God.Ali [00:44:02]: It's literally just an another. It's like, yeah, that's the most simple way to explain it, and it seems like he got trivial speed ups there. But it seems that the complexity with training, it's almost like in our mind at least, it's almost as complex as training GANs. Like it's like a very delicate balance and oftentimes you, it's just but yeah, it's literally speculative decoding on speculative decoding.Vibhu [00:44:21]: Speculative.Ali [00:44:22]: Yeah. We saw this paper.Vibhu [00:44:24]: It's interesting, right?Ali [00:44:24]: Yeah.Vibhu [00:44:24]: I wouldn't even expect it to be very particular to train, I wouldAli [00:44:29]: Right.Vibhu [00:44:29]: The naive part of me is like, okay, train speculative decoder.Ali [00:44:32]: But like, and it makes sense, like the whole idea of speculative decoding is you. It's like, it's like almost like the iPhone auto predict version but for a normal model, right? Like you're just, you're just, generating three tokens and you're like, okay, I'll do prefill on them. And so you save those three turns for your original model. Now your speculative decoder is doing three turns of auto regression, so why not just have an even smaller model?Ali [00:44:53]: The other question there is what are the size of speculators? So say forPhilip [00:44:58]: Right. It's like a billion parameters.Ali [00:45:01]: Like for MiniMax, it's. Yeah. It's like one layer. It's like one 60th of the original model usually.Philip [00:45:06]: Yeah. I think we should do a paper when we get back to the office.Philip [00:45:10]: SpeculativeAli [00:45:11]: SpeculativePhilip [00:45:11]: Decoding.Ali [00:45:13]: No, it's, it does seem like how, when do you stop? But then it also seems like if you're able to train spec-spec decode for instance, right? Like if you're able to have a small model that is accurately predicts what the intermediate speculator is gonna predict, that is able to predict what the original target model's gonna predict, then why not just use that smallest model directly, right?Vibhu [00:45:34]: Yeah. This isAli [00:45:35]: Like it seems likeVibhu [00:45:35]: Adjacent to the routing problem.Ali [00:45:36]: Right.Vibhu [00:45:36]: Yeah.Ali [00:45:36]: Right.Philip [00:45:37]: The thing with speculators is one of the practical constraints on using them is that you do have to run a small model on the same hardware that you're running the big model on. There is a orchestration and resource competition problem inherent in that, and that is one of the constraints on speculation in general, is that draft tokens cost resources to create and cost software complexity to manage. And so if you have like infinitely recursive speculators, you add in quite a bit of that complexity on the actual implementation within the inference engine as well, not just in the training process.Vibhu [00:46:17]: I was gonna say, I would wonder if you could do similar, like distillation and pruning of, it's the same thing, it's just a model. Can we not just distill a lot of the weights, quantize the speculator, out of my domain? The question that also comes up is, this is all for big server workloads, right? How much of this applies to, say I have this MacBook, I wanna run Gemma really efficiently. Similar problems, not the same?Local AI vs. Data Center InferencePhilip [00:46:45]: Pretty different. I talked to Selo, about this on his podcast a couple weeks ago. The difference between inference engineering for the data center and for production workloads versus inference engineering for local AI, is that we start with fundamentally like different constraints and different goals. With local AI, it's how do I fit this model onto my hardware and then make it less dumb? And with data center influence, it's how do I load this model and then make it less slow? And we care about less dumb, and they care about less slow. But the local AI inference engineering ecosystem, I think has a lot for us to learn from in the data center space. They are experts in various forms of quantization, including dynamic quantization that we just don't touch, in the pruning, in the distillation, in the, layer removal. There'Ali [00:47:42]: Layer removal matters less.Philip [00:47:43]: Yeah. There'Ali [00:47:44]: No one loves pruning really.Philip [00:47:45]: Yeah. Well, but the, but they doVibhu [00:47:46]: Which is surprising, right? But that's, that's a whole different thingPhilip [00:47:48]: Just to fit something on the laptop.Ali [00:47:50]: Right.Philip [00:47:50]: So yeah, it's a, it's an interesting, it's an interesting space. Not necessarily that like their techniques make sense for us to do in the data center, because we have different resources and different goals, but more that the process as well as the openness of that field is something to, admire.Ali [00:48:12]: Yeah. Like to your point, like, certain optimizations that would. Like for instance, Turbo Quantum Sharper, like it made such huge hype on that and we did like a whole deep dive on Twitter and like said, what is it? How does it work? Why is it good or not? And it took off and it was implemented on local devices because your memory bandwidth is so slow on like a MacBook, for instance. But try putting the same thing on like an NVIDIA GPU on a B200 Turbo quant would not be. Like, it would not be used. Like, NVIDIA - Like, NVIDIA made it clear that this is not a good optimization, and we've seen it firsthand where the overhead of doing dequantization, quantization of, in the kernel itself with turbo quant kernel, each end is much slower than the time that you save from doing the bandwidth. ‘Cause on the B200s, you have like 3.5 terabytes per second. You don't need decrease the storage that much. You don't need to do, FP4 KV cache. You don't need to use a requant. There's, there's, there's better optimizations to be made. But on Edge devices, it's extremely important, it's extremely useful. So, seems to be, like, different optimizations there, but then they're all uniquely combined with like all you wanna quantize the model, you wanna do speculative decoding, like certain common prefixes with bothPhilip [00:49:18]: Principles.Ali [00:49:19]: Yeah, exactly. Exactly. Exactly.Philip [00:49:20]: They also do a lot of work on, model parallelism, especially over, heterogeneous topology, where you have, some sparks and they are wired together with, Ethernet, DGX sparks.Ali [00:49:35]: Yeah, this is the Exo Labs guys.Philip [00:49:36]: Yeah. You have, a nu

Michael has a limited PTQ coming up in Las Vegas, so we're learning limited! They both independently agreed what the strongest archetype was, but what tips and tricks have they found in the format? *** NEW SPONSOR vaultgamestore.com Use code "GerryT" for 5% off your order or 5% bonus credit on trade-in! *** patreon.com/gerrytpodcast youtube.com/g3rryt twitter.com/g3rryt twitter.com/mwsmtg_ twitch.tv/gerry_t Edited by: Conor O'Donnell (@conorpodonnell) *** Music: Mega Man 2 "Ending theme" Remix by zookun | Music composed by Manami Matsumae & Takashi Tateishi

The CRAZIEST series of events lead to an INSTANT-CLASSIC for the PBA Tour! This week we review the Mike Aulby Nevada Classic! We take a look at two players making it from the PTQ onto the show, then the record setting roll off between EJ Tackett and Ethan Fiore! In the championship match, we saw Andrew Anderson get paybacks against EJ to win the Mike Aulby Nevada Classic! There was a TON of story lines. usagiimports.com code OTG at checkout!

We are coming to you LIVE from the main stage at Magic Con Chicago to give a run down of this jam packed weekend! We debrief with Ethan about his Pro Tour highs and lows, Ben's PTQ run, and then focus up on the game plan going in to the PT and what all went down with Ethan's Day 1 Draft! Use Code "LOL" at CoolStuffInc for 5% off!

Cette semaine Charles propose son 3e top8 de PTQ d'affilée. On vous raconte ses aventures fleuries ! Nouvelle liste Lotus classique https://twitter.com/fireshoes/status/1660348643121983490  UR Turns https://twitter.com/MetropolisCnter/status/1657839342675427330/photo/2  Champion Showcase https://twitter.com/OndrejStrasky/status/1658571423894388741?t=ZtBm77Yq8R2IMbidhuGRbg&s=19  Arena Championship 3 https://twitter.com/val_pl_mafr/status/1661391009782153220  Modern Shadow https://twitter.com/Mengu09/status/1658004400407822337?t=ImX6UVk1_6u3Sfr9FK_1Wg&s=19  Sign Out https://twitter.com/PlayMTG/status/1661138615928324099?t=CG7o2GkiyuhdFLCCijt1iQ&s=19   https://twitter.com/volodarik/status/1657755496852475906  https://twitter.com/PlayStormgate/status/1658962696132521984/photo/1  Intro par In Uchronia  https://www.facebook.com/inuchronia/  https://www.youtube.com/user/inuchronia  Art par Bandit  https://twitter.com/BanditMTG1  Rejoignez notre discord !  https://discord.gg/VtwSyy9  Twitters :  Charles : https://twitter.com/WickedFridge  Théau : https://twitter.com/TheauMery  Twitchs :  Théau : https://www.twitch.tv/inoveletux  Charles : https://www.twitch.tv/wickedfridge 

Large language models (LLMs) show excellent performance but are compute- and memory-intensive. Quantization can reduce memory and accelerate inference. However, for LLMs beyond 100 billion parameters, existing methods cannot maintain accuracy or do not run efficiently on hardware. We propose SmoothQuant, a training-free, accuracy-preserving, and general-purpose post-training quantization (PTQ) solution to enable 8-bit weight, 8-bit activation (W8A8) quantization for LLMs. Based on the fact that weights are easy to quantize while activations are not, SmoothQuant smooths the activation outliers by offline migrating the quantization difficulty from activations to weights with a mathematically equivalent transformation. 2022: Guangxuan Xiao, Ji Lin, Mickael Seznec, Julien Demouth, Song Han https://arxiv.org/pdf/2211.10438v4.pdf

Pro Tour Phyrexia is in the books and we thought we should get the inside scoop from someone who was actually there riding the highs and lows (in equal measure to a perfectly even 8-8 record) - perennial PTQ champion Tommy Ashton. Patreon / Twitter / Discord

The guys debate their thoughts on the latest tour controversy; the whether or not the PTQ system needs a change.  They also update you on the Shawnee Classic, and did Kyle bring home the tree-peat?#bowling #PBA #downlanepodcast Merch: https://www.etsy.com/shop/DownLanePodcastYouTube: https://youtube.com/c/DownLanePodcastFacebook: https://www.facebook.com/DownLanePodcastInstagram: https://www.instagram.com/downlanepod/Twitter: https://twitter.com/DownLanePodcastMailing List: http://eepurl.com/ht8x5fHosts:Kyle HainesAnthony ScacciaProducer:Austin VanBurenMusic: Active Aggression

Thank you for listening to The Brief Case! A podcast for lawyers, hosted by lawyer and cartoonist Sarah-Elke Kraal. Catch us on Instagram (@briefcasepod) and the world wide web: www.briefcasepod.com. My guest in this episode is: Karen Gaston, Barrister, Queensland Bar and Accredited Specialist (Succession Law)—Qld Succession: The Capacity Continuum Karen discusses: Queensland Handbook for Practitioners on Legal Capacity, as prepared by Allens and Queensland Advocacy Incorporated Banks v Goodfellow (1870) LR 5 QB 549 (as adapted for modern life in more recent cases) Ss14, 15 Succession Act 1981 (Qld) Birt v PTQ [2013] QSC 13 at para [94]

RAKDOSCAST 328 – Destaques de ONEMP e Heli comentam mudanças no PTQ, preços e resultados com ONE.Download: https://archive.org/download/rakdos-cast-328-destaques-one/RakdosCast-328-DestaquesONE.mp3Spotify: https://open.spotify.com/show/6HO0FvDaUL1y0SK0WYFmK6Manutenção – (00:05:08)Compra – (00:46:05)Combate – (00:51:54)Principal – (01:05:37)Final – (01:35:34)Links:*Artigo Heliga https://www.ligamagic.com.br/?view=artigos/view&aid=4408 *Arte de SL https://magic.wizards.com/en/news/announcements/update-on-secret-lair-ssssssnakessssss-stonecoil-serpent-art *Mudanças no PT polemicas https://mtgrocks.com/wizards-latest-change-may-be-the-end-of-multiple-mtg-formats/*Lendas de ONE https://magic.wizards.com/en/news/feature/the-legends-of-phyrexia-all-will-be-one *arte myr https://magic.wizards.com/pt-BR/news/announcements/extended-art-urtet-remnant-of-memnarch-rarity *Chupeta ONE https://magic.wizards.com/en/news/announcements/phyrexia-all-will-be-one-jumpstart-booster-themes-and-card-lists *Jace e Ajani completos em 2017 https://www.reddit.com/r/magicTCG/comments/10isvys/in_2017_magic_artist_daarken_was_commissioned_by/ *Deckanpioneiro PT ONe que podem aparecer https://magic.gg/news/metagame-mentor-10-surprising-pioneer-decks-that-can-win-pro-tour-phyrexia?_ga=2.33025534.1942484718.1675523903-229366416.1668767211 *Capa listas https://www.ligamagic.com.br/?view=artigos/view&aid=4413 *Cartas Sanduba: https://www.ligamagic.com.br/?view=artigos/view&aid=4406*Cartas Carol Anet https://www.ligamagic.com.br/?view=artigos/view&aid=4403*Top10 cartas t2 https://www.mtggoldfish.com/articles/phyrexia-all-will-be-one-top-10-standard-cards*Top10 cartas Pioneer https://www.mtggoldfish.com/articles/phyrexia-all-will-be-one-top-10-pioneer-explorer-cardsMusic: Back and Forth - Silent PartnerAdventures by A Himitsu https://soundcloud.com/a-himitsu / Creative Commons — Attribution 3.0 Unported— CC BY 3.0 Rev - EveninglandSupport by RFM - NCM: https://bit.ly/2xGHypMBENSOUND http://www.bensound.com/royalty-free-musicCreative Commons — Attribution 3.0 Unported— CC BY 3.0 Spotify: https://open.spotify.com/show/6HO0FvDaUL1y0SK0WYFmK6Contato: https://rakdoscast.wordpress.com/sobre/Apoie: https://rakdoscast.wordpress.com/formas-de-apoio/

Download RAKDOSCAST 328 – Destaques de ONEMP e Heli comentam mudanças no PTQ, preços e resultados com ONE. Manutenção – (00:05:08)Compra – (00:46:05)Combate – (00:51:54)Principal – (01:05:37)Final – (01:35:34) Links:*Artigo Heliga https://www.ligamagic.com.br/?view=artigos/view&aid=4408*Arte de SL https://magic.wizards.com/en/news/announcements/update-on-secret-lair-ssssssnakessssss-stonecoil-serpent-art*Mudanças no PT polemicas https://mtgrocks.com/wizards-latest-change-may-be-the-end-of-multiple-mtg-formats/*Lendas de ONE https://magic.wizards.com/en/news/feature/the-legends-of-phyrexia-all-will-be-one*arte myr https://magic.wizards.com/pt-BR/news/announcements/extended-art-urtet-remnant-of-memnarch-rarity*Chupeta ONE https://magic.wizards.com/en/news/announcements/phyrexia-all-will-be-one-jumpstart-booster-themes-and-card-lists*Jace e Ajani completos em 2017 https://www.reddit.com/r/magicTCG/comments/10isvys/in_2017_magic_artist_daarken_was_commissioned_by/*Deckanpioneiro PT ONe que […]

The guys discuss the upcoming US Open as well as their experience bowling on last year's PTQ pattern.  #bowling #pba #downlane podcastMerch: https://www.etsy.com/shop/DownLanePodcastYouTube: https://youtube.com/c/DownLanePodcastFacebook: https://www.facebook.com/DownLanePodcastInstagram: https://www.instagram.com/downlanepod/Twitter: https://twitter.com/DownLanePodcastMailing List: http://eepurl.com/ht8x5fHosts:Kyle HainesAnthony ScacciaProducer:Austin VanBurenMusic: Active Aggression

Melissa DeTora is a Senior Game Designer at Wizards of the Coast and currently leads the MTG Casual Play Design team. She is the first woman to Top 8 a Magic Pro Tour. [8:03] Melissa's WOTC journey: Intern -> Senior Game Designer [9:47] What is Play Design? [13:38] Play Design -> Casual Play Design [17:41] Playtesting Commander during COVID-19 [23:26] Designing for "Casual" [25:57] Play Design interactions with Commander RC and CAG [28:21] Too many treasures / Dockside Extortionist is OP [29:36] Arcane Signet [32:06] "I'm still learning things every day" / "I don't want to go to FNM" [37:58] Melissa's mentors [43:21] Design story: Tovolar [46:33] Looking back: Mutate, Companion mechanics [53:46] Looking back: past sets [56:43] Self-criticality as a designer [58:41] Self-criticality as a player [1:00:39] Mindset change: PTQ to PT [1:02:36] "You are probably bad at Magic" [1:05:38] Raphael Levy [1:09:22] Playtesting at the DeTora residence / Joel Larsson, pathfinder [1:13:15] Parents and Magic [1:14:22] Pump It Up [1:24:06] Women in Magic: then vs. now Show notes: humansofmagic.com Patreon: patreon.com/humansofmagic Music: @kuplasound

The boys are back with a show covering a menagerie of topics, including their first impressions of Dominaria United! Also featuring Bryan's PTQ top 8 flameout, how and when to goldfish, recruiting others to join your cult, and the importance of clear communication in Magic. Something for everyone! Music: Mega Man 2 "Ending theme" Remix by zookun | Music composed by Manami Matsumae & Takashi Tateishi arenadecklists.gg patreon.com/arenadecklists twitch.tv/arenadecklists youtube.com/arenadecklists twitter.com/arenadecklists twitter.com/g3rryt twitter.com/bryango

Not only were there 3 major online events for Pioneer this weekend, including an over 300 player PTQ, but we have new Challenger Decks to talk about! So bare with us while Ruckman is still getting over Covid this week. Looking for more pioneer content? Visit: playingpioneer.comWant to support the show? You can find our Patreon here: https://www.patreon.com/crew3mtgSaturday Super Qualifier: https://www.mtggoldfish.com/tournament/pioneer-super-qualifier-12437629#paperSaturday Challenge: https://www.mtggoldfish.com/tournament/pioneer-challenge-12437641#paperSunday Challenge: https://www.mtggoldfish.com/tournament/pioneer-challenge-12437645#paperPioneer League 7-04: https://www.mtggoldfish.com/tournament/pioneer-league-2022-07-04#paperBuy a playmat or used our Inked Gaming affiliate link here: https://bit.ly/3aX4hzOWant to keep up with the show? Join our Discord discord.gg/Wydsb6xcTC or follow us on twitter @Crew3podcastWant more Crew3 content? Check out our YouTube channel or watch our weekly streams on Twitch.If you like the show, please share us with your friends and leave a review! 

Just like old times, the boys are doing their best to get everyone ready for a PTQ season! This particular episode is designed to serve as a timeless guide to the small steps you can take along the way to guide yourself to success. From deck selection tips to evaluating your opponents, there's something here to give everyone the edge they're searching for in securing that (now hypothetical) blue envelope! Music: Mega Man 2 "Ending theme" Remix by zookun | Music composed by Manami Matsumae & Takashi Tateishi arenadecklists.gg patreon.com/arenadecklists twitch.tv/arenadecklists youtube.com/arenadecklists twitter.com/arenadecklists twitter.com/g3rryt twitter.com/bryango

Season 13, Episode 19: Weekly Roundup + Flashback (Fable of the Mirror-Breaker)   We have seen the future of competitive Magic, and it is Regional Championships. The Pro Tour is back in a big way, and with it comes a surprising Pioneer renaissance. The lynchpin of the new system is an event that has not existed before: the Regional Championships, which will feed each Pro Tour three times per year. Part Nationals, part regional PTQ, part Magic Fest, Regional Championships are by far the most exciting component of the new organized play system.    In this episode of Faithless Brewing, Cavedan and Morde analyze the new structure and explain why Regional Championships are a very good thing. We also take a look at six early previews from Streets of New Capenna, and report on our testing results with Fable of the Mirror-Breaker in Modern and Pioneer.   **** Like our content? Support us on Patreon and join our brewing community! **** Decklists for this episode can be viewed at FaithlessBrewing.com **** S13E19 Decklists and Timestamps Roundup: Return of the Pro Tour [4:19] The new system explained [8:17] Pioneer is back [13:17] What are Regional Championships? [18:50] What's missing from the new system [25:28] The verdict New Capenna Previews [31:56] Maestros Charm [37:18] Obscura Charm [42:24] Cabaretti Charm [43:07] Jetmir, Nexus of Revels [44:10] Raffine, Scheming Seer [46:14] Lord Xander, the Collector Flashback: Fable of the Mirror-Breaker [52:20] Mardu Reanimator Blink [1:00:41] David's Mardu Fable Showdown v2

Turns out that Neon Dynasty may have had a little more impact on our formats than we originally anticipated. The reason why? Why, it's (mostly) thanks to Magic's favorite no-downside card type: Artifacts. Tune in as we try out builds of Urzaful/Urzaless Affinity and UW Hammer in Modern, Vehicle Tribal in Historic, and a whole new Ensoul Artifact build in Pioneer. But first, we take a look across 3 high level Modern MTGO events from the weekend. Dave asks for forgiveness. Shane withholds a harsh judgement. Stan gets chipped. The Break Down: Not many NEO cards in the Challenges The Dive Down: NEO cards in Leagues: Vehicles, Affinity, and Ensoul The Dive Down, cont: Stan's New Love: 4-0in' Prelims Check out our NEW SPONSOR, Barrister and Mann! https://www.barristerandmann.com/ Use code THEDIVEDOWN2022 for 15% off your first order of some incredible fragrances, soaps, beard oils, and more. Become a citizen of The Dive Down Nation!: http://www.patreon.com/thedivedown Get 15% off your first 2 months of ManaTraders! https://www.manatraders.com/?medium=thedivedown and use code "THEDIVEDOWN2022" Timestamps: 2:56 - Housekeeping 7:30 - The Break Down begins - Saturday's Challenge with NEW cards! 18:13 - The Sunday Challenge 25:25 - The Modern PTQ 30:25 - Some tournament takeaways 35:34 - A fun diversion 38:10 - The Dive Down begins - S/B/H Artifacts Ed. 40:20 - Shane with Historic Vehicles 57:02 - Dave on Modern Affinity 1:09:44 - Dave on Modern Vehicles 1:18:04 - Stan on Azorius Hammertime 1:31:34 - Stan on Ensoul Artifact 1:40:55 - Oni-Cult Anvil, briefly 1:43:15 - Closing out Links from this week's episode: Saturday's Challenge: https://www.mtggoldfish.com/tournament/modern-challenge-12387446#paper Sunday's Challenge: https://www.mtggoldfish.com/tournament/modern-challenge-12387452#paper Sunday's PTQ: https://www.mtggoldfish.com/tournament/modern-premier-12387425#paper Historic Vehicles: https://twitter.com/jaredfarris/status/1491972071404052481?s=20&t=ITEHIewCd0CHmRVSBee6Xg Spike's Consulate Dreadnaught Affinity: https://www.streamdecker.com/deck/ou3s6zIwy SpiderSpace Affinity: https://twitter.com/SpiderSpaceMTG/status/1492623539773394944?s=20&t=qGj6nPhmn6AhDrLdDAknNw Azorius Hammertime: https://www.mtggoldfish.com/deck/4613330#paper Pioneer Ensoul Artifact: https://www.mtggoldfish.com/deck/4614011#paper Our opening music is Nowhere - You Never Knew, and our closing music is Space Blood - Goro? Is That Your Christian Name? Watch us stream our episodes every Sunday night at 8pm Central: https://www.twitch.tv/thedivedown_shane email us: thedivedown@gmail.com (mailto:thedivedown@gmail.com) twitter: https://twitter.com/thedivedown Leave us an audio message: https://www.podinbox.com/thedivedown

This week we dive into part two of our series in our anticipation of the return to paper Magic — The Road to Vegas. Dave, Stan, and Shane go through the ins and outs of the next tier of Modern's meta. We're looking at two boomer decks — Jund (secretly zoomer?) and UW Control — and one newmer deck in Hammer. Find out which one we think is best positioned for MTG Vegas. Before that we talk about the latest Historic changes, and go through the results of one of last week's MTGO Modern PTQs. Stan controls. Shane hammers. Dave ionizes. The Break Down: Modern Super Showcase WKQ The Dive Down: Boomer Decks The Wind Down: The Short List Become a citizen of The Dive Down Nation!: http://www.patreon.com/thedivedown Get 15% off your first 2 months of ManaTraders! https://www.manatraders.com/?medium=thedivedown and use code "thedivedown2021" Timestamps: 3:33 - Housekeeping 5:36 - Historic Bannings! 10:29 - Historic digital only card changes! 16:13 - The Modern PTQ Top 16 analysis 25:56 - The PTQ meta 34:18 - The Dive Down begins: The Road to Vegas vol.2 36:20 - Dave on Jund Saga 1:05:52 - Stan on Azorius Control 1:29:18 - Shane on Hammertime 1:51:35 - What's on our decks to play shortlist? Links from this week's episode: Modern PTQ Results: https://magic.wizards.com/en/articles/archive/mtgo-standings/modern-premier-2021-10-16?xd2 Moonblade Deck: https://magic.wizards.com/en/articles/archive/mtgo-standings/modern-challenge-2021-10-17?xd2#taliskerstplace Jund Saga: https://www.mtggoldfish.com/deck/4325030#paper UW Control: https://magic.wizards.com/en/articles/archive/mtgo-standings/modern-premier-2021-10-16 Hammer: https://www.mtggoldfish.com/deck/4109042#paper Our opening music is Nowhere - You Never Knew, and our closing music is Space Blood - Goro? Is That Your Christian Name? Watch us stream our episodes every Sunday night at 8pm Central: https://www.twitch.tv/thedivedown_shane Leave us an audio message in our PodInbox: https://www.podinbox.com/thedivedown email us: thedivedown@gmail.com (mailto:thedivedown@gmail.com) twitter: https://twitter.com/thedivedown

About six months ago, we took a break from our normal programming to refresh our brains by asking some questions about the other person's Magic career.  We're taking another one this week, the topics of discussion include the worst punt in a PTQ, what song is playing when brewing, and what decks shaped our early Magic memories.  Enjoy!

On this week's epsiode, MTGO grinders Bryan (Veveil) and Charlie (StryxFamiliar) make a return to the podcast to talk about their experiences in KHM with a focus on Sealed to prep you for the Arena qualifier.  LLU Discord: bit.ly/jointhedischord Bryan's PTQ deck: https://twitter.com/Veveil/status/1360475823095095296 Charlie's PTQ deck: https://twitter.com/stryxfamiliar/status/1360481102306115585

Faithless Brewing, Episode 79: Leveling Up with Aspiringspike Fresh off a PTQ win, Evart Moughon aka Aspiringspike returns to answer listener questions about the state of Modern and Pioneer, the future of control, tips for aspiring brewers, and how to improve as a Magic player. Also on the Flashback: another trophy with Zombies, and a cool new twist on Reanimator. Where to find EvartStream: twitch.tv/aspiringspike (9am-3pm CST, Mon-Fri)Twitter: @MoughonEvart Decklists for this episode can be found at our new home page, FaithlessBrewing.com. Visit us for articles, bonus decklists, and more!  New articles this week:"At the Dive-In: Heliod Company Is the Deck to Beat This Weekend" by Stanislav Like our content? Become a supporter and join the Faithless Family: patreon.com/faithlessbrewing

Faithless Brewing, Episode 79: Leveling Up with Aspiringspike Fresh off a PTQ win, Evart Moughon aka Aspiringspike returns to answer listener questions about the state of Modern and Pioneer, the future of control, tips for aspiring brewers, and how to improve as a Magic player. Also on the Flashback: another trophy with Zombies, and a cool new twist on Reanimator. Where to find EvartStream: twitch.tv/aspiringspike (9am-3pm CST, Mon-Fri)Twitter: @MoughonEvart Decklists for this episode can be found at our new home page, FaithlessBrewing.com. Visit us for articles, bonus decklists, and more!  New articles this week:"At the Dive-In: Heliod Company Is the Deck to Beat This Weekend" by Stanislav Like our content? Become a supporter and join the Faithless Family: patreon.com/faithlessbrewing

Faithless Brewing, Episode 79: Leveling Up with Aspiringspike Fresh off a PTQ win, Evart Moughon aka Aspiringspike returns to answer listener questions about the state of Modern and Pioneer, the future of control, tips for aspiring brewers, and how to improve as a Magic player. Also on the Flashback: another trophy with Zombies, and a cool new twist on Reanimator. Where to find EvartStream: twitch.tv/aspiringspike (9am-3pm CST, Mon-Fri)Twitter: @MoughonEvart Decklists for this episode can be found at our new home page, FaithlessBrewing.com. Visit us for articles, bonus decklists, and more!  New articles this week:"At the Dive-In: Heliod Company Is the Deck to Beat This Weekend" by Stanislav Like our content? Become a supporter and join the Faithless Family: patreon.com/faithlessbrewing

Cedric Phillips unveils his Top 5 topics for Monday. November 2: 5. NFL Week 8 results and recap (1:35) 4. More Nick Cage movies and trailers have been consumed (10:20) 3. Winning a PTQ with Boros Wizards on Saturday (17:44) 2. Mindsets in Magic and why I disagree with them (24:55) 1. The 2020 Election and panic attacks (36:40) Enjoy!

Thanks to AspiringSpike for joining us! https://twitter.com/MoughonEvart https://www.twitch.tv/aspiringspike This week is all about AspiringSpikes 2 top finishes on MTGO, including a PTQ win! Also with Gab making the finals of the MagicFest Online Finals he also had some Standard to go over. Enjoyed the episode? Don't forget to follow the Podcast and our Social Media: Patreon: https://www.patreon.com/MidweekMetagame Twitter: https://twitter.com/MidweekMetagame Gab: https://twitter.com/gabnassif https://twitch.tv/yellowhat Pat: https://twitter.com/ghett_smart Harry: https://twitter.com/harrymtg

Sometimes things happen, but the show does go on! PTQ has a new name... "On the DF"! It may be a new show name, but Darren is still singing away while he brings you entertainment news that we can all use right now!"On the DF" is a Branas Enterprises production.BE the Voice Podcast NetworkWebsite: branasenterprises.comFacebook/Instagram/Twitter/TikTok: branasent

This week the Beans discuss their performance (or lack therof) at the recent Good Games Premier series PTQ. They also touch on the recent Banned & Restricted announcement, how their Arena league is going and whats going on in the competitive scene in the next couple of weeksCrackers Selesnya Auras deck can be found hereEnjoy!Link to our Discord server for all to join - https://discord.gg/NuvSY9KFollow us at:twitter.com/MagicBeansCastfacebook.com/magicbeanscastEmail us at magicbeanscast@gmail.comOn Twitter we are:Shorty - @pieceincCracker - @joelhill_Chewy - @chewyMTGScott- @therealblastaChris - @pollywafflemtg

Nous avons passé un super week-end à Lyon, et avons tous les deux cash ! Dans cette épisode on parlera de Pioneer, de nos résultats en PTQ, ainsi que de standard. On parlera aussi des bans, et de toutes nos anecdotes de Lyon, ainsi qu'un petit point plaines sur Sram Aura !Intro et Outro par In Uchronia https://www.facebook.com/inuchronia/ https://www.youtube.com/user/inuchronia Art par Safia Loucif https://www.instagram.com/loucifsafia/ Rejoignez notre discord ! https://discord.gg/VtwSyy9 Tweet de CFB sur la decklist d'Alain Bardini https://twitter.com/ChannelFireball/status/1237112540238397441 Metagame du GP Lyon https://www.channelfireball.com/all-strategy/articles/the-standard-metagame-at-grand-prix-lyon/ https://twitter.com/ChannelFireball/status/1236674745627750400 Ban announcement https://magic.wizards.com/en/articles/archive/news/march-9-2020-banned-and-restricted-announcement Twitters : Charles : https://twitter.com/WickedFridge Théau : https://twitter.com/TheauMery Twitchs : Théau : https://www.twitch.tv/inoveletux Charles : https://www.twitch.tv/wickedfridge 

In the third episode of MTG Standard Podcast, Jason and Jana talk about the standard meta-game at SCG Philadelphia and PTQ at MagicFest Phoenix. In both tournaments, Azorius Control dominates the scene. Is Azorius Control truly the best deck of the current standard format? MTG Standard Podcast analyzes this blue-white powerhouse (and other archetypes) in more detail. Jason and Jana also make a prediction on deck archetypes that may make an appearance at Magic Championship Worlds in the upcoming weekend

Intro Magic has had a major security breach and we want to run down the archetypes and what kind of two-step authentication steps we can attempt to stop them. Discuss Card and what it brings to eternal formats The first weekend culprits (legacy) MARTINMEDMITTEN (5TH PLACE) ARONGOMU (10TH PLACE) THREPIO (20TH PLACE) Max Breach Control https://www.mtggoldfish.com/archetype/legacy-ubrg-158554#paper Delver Breach Legacy Premier Event No Legacy Gp’s the first half of the year https://magic.wizards.com/en/articles/archive/mtgo-standings/legacy-premier-2020-01-19 https://magic.wizards.com/en/articles/archive/mtgo-standings/legacy-challenge-2020-01-20 Fun ofs Echo Monk Narset Legacy Challenge Svaca wrecking with Klothys in challenge Doomsday is back???? (prob not) Oko Landstill Breakfast!!!! Artificer's Intuition Negator 77-The difficulty some players had being able to play in the PTQ last weekend and the upcoming showcase event due to the reduction in league QP prizes relative to the forced entry for those events (40QP), when it was 35 for the "playoff" events last year and open entry (P Points/Tix) for the PTQ's. People almost have to grind a ton or find time for the awkwardly scheduled daily (ahem sorry, prelim) events. so spike a prelim or GRIND a lot 4 for a 5-0, 2 for a 4-1.. none for 3-2 so 10 5-0's, 20 4-1's or a combo I won the trophy race and barely had the 80 required for the ptq and this weekends showcase playoff Pareon Shoutout Cody W. Social Media Steve https://twitter.com/raistlinim Lawrence https://twitter.com/LawrenceHarmon Liz https://twitter.com/elioftheveil ThirstforKnowledgeCast https://twitter.com/ThirstForCast

Neste episódio falamos sobre os resultados do PTQ and Legacy Challenge online, algumas das cartas de Theros Beyond Death que ja tiveram um impacto no formato e sobre o calendario dos GPs. Legacy Challenge Legacy PTQ(MTGO) Emperor of Legacy Tournament 8th Season @ Hareruya (Japan) Fausto http://eternalmagic.com.br/ Romário Follow @romarioneto3 Twitch YouTube

On Episode Eighty-Six of Fishin, Kody and Dorian are joined by Cody Hope who recently top eighted a pioneer PTQ with Merfolk! We discuss his list, primary strategies, as well as our hopes for the future of Merfolk in the format.Find the Show here:@iHeartRadio: https://ihr.fm/2JlWzQ0@GooglePlay: https://bit.ly/2q6JRMh@ApplePodcasts: https://apple.co/2Gy7NmS@spreaker: https://bit.ly/2GCMhcA@Stitcher: https://bit.ly/2JlQOSj@Spotify: https://spoti.fi/2LmGsGJ@youtube: https://bit.ly/2ssGp3ZJoin our discord, The Wanderwine Hub, to discuss all things merfolk!https://discord.gg/Xzt9QBYIf you have found our show to be helpful and worthwhile, please consider donating as much as a dollar a month to our Patreon! We have sweet rewards starting at just the 2 dollar level: https://bit.ly/2q5LbisWe are sponsored by @inkedgaming! Use this link to get 10% off and help support the show! inkedgaming.com/fishcastmtgGet your merfolk tokens from @cardamajigs here:https://cardamajigs.com/collections/fishin-a-merfolk-podcastCheck out our stream. www.twitch.tv/fishcastmtg Twitter handles:Fishin': @fishcastmtg -instagram and facebook too!Kody: @notkodysmith Matt: @MatthewCaudill8 Dorian: @krinklebiscuitemail: fishcastmtg@gmail.comPodcast cover art credits go to to Tessa and Hunter Pruitt. Stream/logo/ emotes art and overlay by @ishton on twitter, and Intro video by @jakebossmtgFishin: A Merfolk Podcast is unofficial Fan Content permitted under the Fan Content Policy. Not approved/endorsed by Wizards. Portions of the materials used are the property of Wizards of the Coast. ©Wizards of the Coast LLC.

On Episode Eighty-Six of Fishin, Kody and Dorian are joined by Cody Hope who recently top eighted a pioneer PTQ with Merfolk! We discuss his list, primary strategies, as well as our hopes for the future of Merfolk in the format.Find the Show here:@iHeartRadio: https://ihr.fm/2JlWzQ0@GooglePlay: https://bit.ly/2q6JRMh@ApplePodcasts: https://apple.co/2Gy7NmS@spreaker: https://bit.ly/2GCMhcA@Stitcher: https://bit.ly/2JlQOSj@Spotify: https://spoti.fi/2LmGsGJ@youtube: https://bit.ly/2ssGp3ZJoin our discord, The Wanderwine Hub, to discuss all things merfolk!https://discord.gg/Xzt9QBYIf you have found our show to be helpful and worthwhile, please consider donating as much as a dollar a month to our Patreon! We have sweet rewards starting at just the 2 dollar level: https://bit.ly/2q5LbisWe are sponsored by @inkedgaming! Use this link to get 10% off and help support the show! inkedgaming.com/fishcastmtgGet your merfolk tokens from @cardamajigs here:https://cardamajigs.com/collections/fishin-a-merfolk-podcastCheck out our stream. www.twitch.tv/fishcastmtg Twitter handles:Fishin': @fishcastmtg -instagram and facebook too!Kody: @notkodysmith Matt: @MatthewCaudill8 Dorian: @krinklebiscuitemail: fishcastmtg@gmail.comPodcast cover art credits go to to Tessa and Hunter Pruitt. Stream/logo/ emotes art and overlay by @ishton on twitter, and Intro video by @jakebossmtgFishin: A Merfolk Podcast is unofficial Fan Content permitted under the Fan Content Policy. Not approved/endorsed by Wizards. Portions of the materials used are the property of Wizards of the Coast. ©Wizards of the Coast LLC.

We're back after a week off for what will be our final episode of 2019. This week we're talking about the banning of Nexus of Fate and Oko, Thief of Crowns (RIP the twink with the most abs). As well as looking at some of the recent PTQ and League Results. Please be sure to leaves us a review and follow us on Twitter and Instagram @Crew3Podcast to keep up to date with us. We'll be back in two weeks, so from all of us at Crew 3 enjoy your holidays and have a wonderful new year!

For Part 2, Michael J Flores showed up with cake from Carlos's bakery in Hoboken to celebrate his PTQ win -- more than 23 years after his first PTQ win -- as is the tradition in NY. Topics included the PreTQ, PTQ, and what makes a good Magic format. Followed by some spoilerific discussion of the season finale of WATCHMEN as well as the first three parts of CRISIS ON INFINITE EARTHS. Simic Flash by Michael J Flores 2 Aether Gust 2 Brazen Borrower 2 Mystical Dispute 1 Negate 3 Quench 3 Sinister Sabotage 4 Frilled Mystic 2 Hydroid Krasis 4 Growth Spiral 4 Nightpack Ambusher 4 Nissa, Who Shakes the World 3 Paradise Druid 4 Breeding Pool 2 Castle Vantress 2 Fabled Passage 7 Forest 7 Island 4 Temple of Mystery Sideboard 1 Ugin, the Ineffable 2 Aether Gust 2 Mystical Dispute 2 Hydroid Krasis 4 Cavalier of Thorns 2 Lovestruck Beast 2 Shifting Ceratops --- Support this podcast: https://anchor.fm/top8magic/support

Faithless Brewing, Episode 34: Fires of Invention "If it's free, it's me." The timeless words of Bryan Gottlieb teach us to pay close attention to anything that can be cast for free, one of the most powerful and potentially broken effects in all of Magic. Fires of Invention grants that text to any spell in your deck, allowing you to cast it without tapping lands. And then, it does it again. Two free spells every turn, and you still have all your mana available to spend on other effects? Yeah, that's me. Sign us up for a playset! Fires of Invention is Throne of Eldraine's entrant into the "broken four-mana enchantment engine" category, and it has already proven its chops in Standard as well as Pioneer and Modern. More similar to a Wilderness Reclamation than an Experimental Frenzy, Fires offers an insane mana advantage, provided that you can successfully navigate its numerous constraints. This week the crew has some explosive builds designed to do just that. You want decklists? "Just tell us how many you want and get out of the way!" Roundup: Weekend Results  Simic Food Ramp (Gerry Thompson) Gruul Kiora (Dan): 3-3, SCG IQ Newington Azorius Approach (Tyler Surovy): 3rd place, SCG IQ Newington  Flashback: Crested Sunmare  Modern GW Eldritch Horse: 1-4 league  Pioneer Orzhov Horse: 3-2 league Abzan Horse: 7-3 leagues  Brew Session: Fires of Invention Modern Sketch: Gruul Fires Escape  Pioneer Sketch 1: Temur Elemental Fires Sketch 2: Niv-Mizzet Fires  Reference lists: StandardJeskai Fires (Paulo Vitor Damo da Rosa)5c Niv Fires (Ken Yukuhiro)5c Golos Fires (Shota Yasooka)  ModernIzzet Turns (lsv, 5-0 league)Izzet Turns (Gfrpaac, 5-0 league)4c As Firetold (PetyrBaelish, 7-2 PTQ)4c As Firetold (Runeskjold, 5-0 league) PioneerJeskai Fires Superfriends Lawson Zandi (Niv-Mizzet streams): https://www.twitch.tv/zanman1414  Contact Us  If you like our show, be sure to join our Patreon and leave us a review! You can do this from the Apple Podcasts app, or from iTunes if on a computer. Thank you for your support! Patreon: https://www.patreon.com/faithlessbrewing Twitter: @FaithlessMTGEmail: faithless.brewing@gmail.com Homepage: faithlessbrewing.podbean.com 

This week Al's out, and Dave Murphy AKA DaveSea89 is in! Dave joins us to discuss Standard, in which Dave recently won a PTQ, Pioneer, in which we accurately predicted the bans, and Secret Lair! Plus a whole lot of smack talk.Check out Dave's WCW vs NWO podcast on Soundcloud.Head on over to Inked Gaming to get yourself a custom playmat, and take advantage of 10% off your entire order with code SKULLCRAIC at checkout!Tweet us @SkullcraicCheck us out on FacebookEmail us at skullcraicpodcast@gmail.com"Skullcraic Theme" composed and performed by Barry Cannon.Barry's YouTubeBarry's Twitter

Dans cet épisode, Théau et Charles nous annoncent leur participation au coverage français officiel du MC VII! Charles fera un retour sur son GP Bologne, qui fut une performance... inédite. On parlera également des résultats des PTQ en Pioneer, ainsi que des bans avec un format qui va aller en se stabilisant. Théau nous parlera de sa qualification à la Palm, et on terminera avec l'Eternal Week-end le week-end du 21 décembre.Enfin, des petites histoires du GP Bologne termineront ce programme. Et comme d'habitude une petite question règles sur la fin.Intro et Outro par In Uchronia https://www.facebook.com/inuchronia/ https://www.youtube.com/user/inuchronia Montage par Romainhttps://twitter.com/Romhappy1Art par Safia Loucifhttps://www.instagram.com/loucifsafia/ Decklists disponibles pour le momenthttps://twitter.com/jpball5/status/1200143159776419841 Twitters :Charles : https://twitter.com/WickedFridge Théau : https://twitter.com/TheauMery 

After Ricky's computer died mid recording Chris and Ruckman decided to start over and record a shorter episode this week covering some of what was discussed in the now lost version of episode 3. Mono-Black's firmly on top after the first two days of MTGO's pioneer week PTQ's how should you combat it? Listen and find out.

This week, we dive into fashion.  Well, mainly, why you should put on pants and leave the house to play Magic.  We talk about our last PTQ and then talk about some hot new standard decks that are floating around the internet.  The main topic is week is how to get ready to play paper magic.  What are the things that you are going to need and how to get them.  We follow up with some Arena and end on the new product Secret Lair. PTQ for PT Phoenix - 1:50 New Standard Decks - 17:55 Part 1: How to play in paper - Where to Play - 36:33 Part 2: How to play in paper - Accessories - 53:12 Part 3: How to play in paper - Getting a Deck 1:11:09 Arena - 1:22:46 Secret Lair Reveal - 1:30:39 Facebook: Casual Try Hard MTG Twitter: @casualtrypod Email: casualtryhardmtg@gmail.com You can find us on the Apple podcast app, Google Play, Podbean, Soundcloud, Spotify, Stitcher or YouTube just search casual try hard. Music by Juan Rodriguez II ZeeManlove.com  

In this episode, Rin and Kyle talk about the latest in meta shake ups from the PTQ, Challenge, and Kyle’s

In this episode, Rin and Kyle talk about the latest in meta shake ups from the PTQ, Challenge, and Kyle’s

First Pick: GOOGLE HANGOUT THIS SUNDAY - 19TH - 10PM ET   Reaction to Helene interview / WMCQ updates: https://twitter.com/HeleneBergeot/status/831878045283684352   Winner of Nationals not "National Champ" Beuhler fixed: just let the Team Captain play, if TC makes Top 2, let 3rd/4th playoff   http://magicorganizedplay.tumblr.com/post/157244798901/grand-prix-sunday-ptq-formats   "Therefore, we decided that PTQs would be Standard for all Limited Grand Prix and Limited for all Constructed Grand Prix."   PTQ cost at GPs -- $50? $75? 225 single flight vs 8x 32-players flights (256 players)   4 new / returning programs for Amonkhet: http://wpn.wizards.com/en/article/whats-new-amonkhet https://twitter.com/misterorange/status/831148246697463813   Gather the Townsfolk: http://markrosewater.tumblr.com/post/157124759613/no-a-lot-of-people-play-modern-we-have-no-plans#notes   Going Infinite https://www.reddit.com/r/magicTCG/comments/5u7lft/the_mtg_documentary_you_want_made/   Red Zone Why Diverse Formats Suck: http://www.hipstersofthecoast.com/2017/02/diverse-formats-suck/   Splash Damage https://www.reddit.com/r/hearthstone/comments/5u0p5y/mtg_even_with_hearthstone_on_google_trends_for/   Pokemon mapped: https://www.reddit.com/r/videos/comments/5u1og8/pokemon_has_started_mapping_their_booster_boxes/   The Finisher Yesterday was Valentine's Day. And you know we all love each other, but I thought it might be nice if we shared that love on air. So we've all prepared some Magic Mics poems to express just how much we like each other.

Cheating, PTQ deckbuilding, SPOILERS!

Cheating, PTQ deckbuilding, SPOILERS!

After a brief discussion about being the biggest idiot ever, the dorks head right into the Magic 2014 cards they showed the freakin' night after we recorded the last episode. It's just one per color, so it doesn't take that long.  ZOMG Fish Illusion!Then Mike and Brian tell us all about a PTQ they went to at Atomic Empire in Durham North Carolina. It's a great shop, you should check it out.  If that's not funny to you, you need to listen to more Cluze. Just sayin'.  Non-spoiler: Mike finished a game in such an amazing way that I can't even bring myself to type it here. Just listen.After a musical break, it's time for the topic Chewie has been trying to do for a month now.  We're taking a look at the various key elements of the guilds from the original Ravnica block and comparing them to the same elements in Return to Ravnica in an effort to determine which one is just, well, better.  This episode we're focusing on the hippies of the Selesnya guild.  So what's better, old weed-smoking hippies or new hemp-wearing hippies?  Let us know what you think!Magic 2014 Card Image Gallery: http://www.wizards.com/magic/tcg/article.aspx?x=mtg/tcg/magic2014coreset/cigRed Hexproof deck Mike liked so much: http://www.wizards.com/magic/magazine/Article.aspx?x=mtg/daily/deck/1251Nostalgia Critic videos: http://thatguywiththeglasses.com/videolinks/thatguywiththeglasses/nostalgia-criticIntro & Outro Music – Diamond by Swift – http://www.myspace.com/swiftband Break Music – No Love by Eminem (ft. Lil Wayne) – http://www.eminem.com This episode's post: http://themanapool.com/podcast/episode-282 Forum thread for this episode: http://cardshark.freeforums.org/episode-282-old-vs-new-hippie-edition-t928.html Check out our archive at http://themanapool.com/category/podcast Our Facebook page: http://www.facebook.com/TheManaPool Our Google+ page: http://gplus.to/themanapool Email us at dorks@themanapool.com Follow us on Twitter at @TheManaPool Cardshark Member Promotions: http://www.cardshark.com/Promotions/Member-Promotions.aspx Subscribe to our RSS feed: http://feeds.feedburner.com/themanapool

In this episode, Mike gets distracted by something early on. But that's not the actual content of the show! Brian and I kick it off by each opening a pack of Zendikar and going through the contents. Then we switch gears and talk about Brian's PTQ experience from a few weekends ago. He did really well. His deck list is in the forum thread for this episode, give it a look and let us know what you think. And of course, we had to go over the freshly released lists for Duel Decks: Garruk vs. Liliana. Rancor! Mutilate! Beasts! Yay! We go over each list, giving a pretty detailed rundown on how we think each one will work. Then of course we have to pick a winner. Keep listening to find out who we thought might come out ahead this time. Music for this episode is freakin' great, you should go check it out! Carl Sagan - A Glorious Dawn ft. Stephen Hawking by John Boswell Youtube video: http://www.youtube.com/watch?v=zSgiXGELjbc Check out more Symphony of Science: http://www.symphonyofscience.com Forum thread for this episode: http://cardshark.freeforums.org/episode-104-ptq-results-duel-decks-and-olive-oil-t421.html Check out our archive at http://themanapool.libsyn.com Email us at themanapool@cardshark.com Follow us on Twitter at http://twitter.com/themanapool Visit Cardshark! http://www.cardshark.com Cardshark Member Promotions: http://www.cardshark.com/member_promotions.asp Subscribe to our RSS feed: http://feeds.feedburner.com/themanapool

Can we say that?This time we've got a lot to go over.  We're Dirkless again due to his recent computer problems, but we're joined by my old roommate Corey again, so that evens out.  We've got a mess of feedback to go over first.  Then Brian and I have to tell you about the recent PTQ we attended.  We ran the decks that were being tested in the last episode, just so you know.  No last minute deck swaps for us.Then we get to the true purpose of this episode.  We're discussing multiplayer deckbuilding.  More specifically, we're going to focus on the differences between building a deck to take on one person and building a deck to handle multiple opponents.  Of course, we sprinkle liberally with our own personal experiences and knowledge.  And there's plenty of giggling, joking, tangential nonsense, and picking on each other, as per usual.  What sort of episode would it be without all that?Forum thread for this episode: http://cardshark.freeforums.org/episode-84-can-we-say-that-multiplayer-deckbuilding-t344.htmlCheck out our archive at http://themanapool.libsyn.comEmail us at themanapool@cardshark.comSend me an article at submissions@cardshark.comCardshark located at http://www.cardshark.comCardshark Member Promotions: http://www.cardshark.com/member_promotions.aspSubscribe to our RSS Feed: http://feeds.feedburner.com/themanapool

This is probably not what you were expecting, is it?  Due to recording time problems, it was either this or no show at all.  And since I know what happens to you rabid Mana Pool fans out there when you don't get your weekly fix, Brian and I came up with a solution.  Let us know if you thought this was good, bad, terrible, stupid, didn't care, greatest thing ever, or whatever you think.  Please give us some feedback here, we're curious. What you're hearing here is us testing for the PTQ that happened this past weekend.  Some people requested that we try playing a game or more on the air, so we decided to give it a shot, killing two birds with one stone.  Brian's playing a Bant Aggro - Finest Hour deck, and I'm playing a Jund Aggro - Cascade deck.  We'll hit you with how the decks performed at the PTQ next episode, as well as get back to our multiplayer shows.  Music for this episode: Respect the Blade by SwiftCheck out Swift at http://www.swift-band.com Forum thread for this episode: http://cardshark.freeforums.org/episode-83-testing-an-experiment-t340.htmlCheck out our archive at http://themanapool.libsyn.comEmail us at themanapool@cardshark.comSend me an article at submissions@cardshark.comCardshark located at http://www.cardshark.comCardshark Member Promotions: http://www.cardshark.com/member_promotions.aspSubscribe to our RSS Feed: http://feeds.feedburner.com/themanapool