Video yükleniyor...

Video Yüklenemedi

Ana Sayfaya Dön

The new DeepSeek V4.1 Flash model is mindblowing - back on top of the open-source model leaderboard and extremely cheap. It has a lot of very smart ways to be efficient and highly capable so I made a video of the forward pass to give you a view of...

Thom_Wolf's profile picture

Thomas Wolf

131,622 subscribers

58,020 görüntüleme • 1 gün önce •via X (Twitter)

Bilim & Teknoloji
LIVE

Anya Rossi• Live Now

Private livecam show

17 Yorum

AionCloud profil fotoğrafı
AionCloud1 gün önce

Beautiful pass. Still a run. Still hours.

Neil profil fotoğrafı
Neil1 gün önce

It's almost impossible to keep up with the local AI world nowadays. It's moving at an insane speed.

Mehdi Zare profil fotoğrafı
Mehdi Zare1 gün önce

I'm watching KV-cache pressure more than raw tokens per second. Better cache reuse is what decides whether parallel coding sessions feel cheap in practice.

Shesaidmewakeup profil fotoğrafı
Shesaidmewakeup1 gün önce

open weights the same day as the drop is the part labs hate.

Information Machine profil fotoğrafı
Information Machine1 gün önce

An MLX version would be nice.

Nick profil fotoğrafı
Nick1 gün önce

The forward-pass view is useful, but the production constraint is memory traffic, not only FLOPs. I’d want the same breakdown with KV-cache size and tokens/sec at long context; that’s where cheap inference gets less cheap.

Marius Laurusevicius profil fotoğrafı
Marius Laurusevicius1 gün önce

The scaffold table is the interesting part: DeepSWE v1.1 runs 65.5 to 74.2 across harnesses and the headline 74.2 is mini-SWE, while Terminal-Bench 2.1 runs 84.1 to 90.6 and the headline 90.6 is DeepSeek's own harness.

John Rood profil fotoğrafı
John Rood1 gün önce

this should be the default model release artifact. benchmarks tell you where it won. an animated forward pass tells builders where the cost went.

Gentleman Tech Bro profil fotoğrafı
Gentleman Tech Bro1 gün önce

Do we know if this has anything approaching an internal world model? I believe Astra is moving this way.

Naveen Saradhi profil fotoğrafı
Naveen Saradhi1 gün önce

The interesting part isn’t just that DeepSeek is back on top. It’s how they’re getting this much capability at this price.

安叫兽|Bird🕊️ 🔶 BNB profil fotoğrafı
安叫兽|Bird🕊️ 🔶 BNB1 gün önce

前向传播可视化比单看榜单直观多了。

MOHAMMED ISMAIL profil fotoğrafı
MOHAMMED ISMAIL1 gün önce

DeepSeek keeps pushing the boundaries of efficiency. ���� Cheap + highly capable open-source models are a huge win for developers. 🤖🚀

David Moore profil fotoğrafı
David Moore1 gün önce

We've seen this pattern before, but closer to home. Thus, whether east or west, the play--whatever it is--I strongly suspect is not in our favor. That is wisdom. If you can accept it, stop using cloud models.

Tanjiro profil fotoğrafı
Tanjiro1 gün önce

So fast and efficient!

Srinivas Avula✨ profil fotoğrafı
Srinivas Avula✨1 gün önce

Let's connect Thomas😊

Xman profil fotoğrafı
Xman1 gün önce

Open weights plus this forward-pass view make the efficiency claims much easier to inspect

VastPlan profil fotoğrafı
VastPlan1 gün önce

开源模型更新，现在连 forward pass 都做成短视频了。可读性本身成了发布资产：别人看懂你怎么省算力，比再刷一分榜更容易被采用。

Benzer Videolar

deepseek v4.1 flash vs secret model tested both with the same prompt at the highest reasoning available > v4.1 flash took 2 hours to finish both tests and cost $2.60 > secret model finished both in 30 minutes, can’t reveal cost yet v4.1 flash spent a lot of time overthinking and running random tests, which slowed it down thanks AIHubmix for providing api access so i could test v4.1 flash hoping that team will cut down the overthinking before the full release, but for now here are the results which model did better here?
0:41

deepseek v4.1 flash vs secret model tested both with the same prompt at the highest reasoning available > v4.1 flash took 2 hours to finish both tests and cost $2.60 > secret model finished both in 30 minutes, can’t reveal cost yet v4.1 flash spent a lot of time overthinking and running random tests, which slowed it down thanks AIHubmix for providing api access so i could test v4.1 flash hoping that team will cut down the overthinking before the full release, but for now here are the results which model did better here?

J A Z I I

28,586 görüntüleme • 3 gün önce

deepseek-v4.1-flash I'm really curious to understand the architecture behind the scenes. 🤔 Model is really fast as you can see in the video! I've got an average decoding speed of around 350 t/s. BTW it is clearly experimental and: - currently, the model is under heavy load and failing quite often. - it thinks like there is not tomorrow - it keeps failing editing file - after 4M tokens my frogger game is still in progress - it can't see images out of the box on DeepSeek Harness, you have to add these config lines - id: deepseek-v4.1-flash-expires-on-0910 name: deepseek-v4.1-flash-expires-on-0910 contextWindow: 1000000 inputModalities: - text - image imagePixelBudget: 640000 imageMaxBytes: 1048576 So play with it, but be aware of the very early stage.
0:22

deepseek-v4.1-flash I'm really curious to understand the architecture behind the scenes. 🤔 Model is really fast as you can see in the video! I've got an average decoding speed of around 350 t/s. BTW it is clearly experimental and: - currently, the model is under heavy load and failing quite often. - it thinks like there is not tomorrow - it keeps failing editing file - after 4M tokens my frogger game is still in progress - it can't see images out of the box on DeepSeek Harness, you have to add these config lines - id: deepseek-v4.1-flash-expires-on-0910 name: deepseek-v4.1-flash-expires-on-0910 contextWindow: 1000000 inputModalities: - text - image imagePixelBudget: 640000 imageMaxBytes: 1048576 So play with it, but be aware of the very early stage.

Ivan Fioravanti ᯅ

26,971 görüntüleme • 3 gün önce

Big day for open source. MiniMax Music 3 is undeniably a state of the art open weights Music Generation model and a real alternative to Suno. My favorite part about open weights: the best is yet to come. Once the community starts playing with the model and training LoRAs, this model only gets better and allows for more control.
1:59

Big day for open source. MiniMax Music 3 is undeniably a state of the art open weights Music Generation model and a real alternative to Suno. My favorite part about open weights: the best is yet to come. Once the community starts playing with the model and training LoRAs, this model only gets better and allows for more control.

rob - comfyui

320,517 görüntüleme • 28 gün önce

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.
1:57

The DeepSeek-R1 paper is a gem! Highly encourage everyone to read it. It's clear that LLM reasoning capabilities can be learned in different ways. RL, if applied correctly and at scale, can lead to some really powerful and interesting scaling and emergent properties. There is more to RL than meets the eye! Here is my breakdown of the paper along with a few tests: The multi-state training might not make sense initially but they provide clues on optimizations that we can continue to tap into. Data quality is still very important for enhancing the usability of the LLM. Unlike other reasoning LLMs, DeepSeek-R1's training recipe and weights are open so we can build on top of it. This opens up exciting research opportunities. About the attached clip: the previous preview model wasn't able to solve this task. DeepSeek-R1 can solve this and many other tasks that o1 can solve. It's a very good model for coding and math.

elvis

141,023 görüntüleme • 1 yıl önce

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!
4:28

Alright guys, wow. Qwen 3.8 27B is basically Qwen 4. What we have here is one of the best one-shot shark survival games I've made with any model. I ran this same prompt through fable, and the biggest thing all models struggle with in this test is the top-down view of the sharks. Fable even had all of the fins reversed and details jumbled on the first turn. Before now, it was really a 2 to 8 turn test to get something final regardless of model. Qwen 3.8 27B just nailed this in one output, no harness at all, just purely the model thoroughly thinking before outputting the final result. On every asset used, different shark variants, everything, it all looks incredibly tight. The boat movement, shark speed and engagement, everything is so much better than I expected, and I was expecting a lot. Imagine what 8 turns would do in a harness? This confirms to me that while Qwen 3.8 27B thinks a lot, for the first time in a local model on the base weights, IT MIGHT ACTUALLY BE WORTH IT This is a Frontier Lab model result, and it ran on my 5090 at 60tps in Q5_K_M. I am just in complete astoundment at what Qwen has accomplished here, and I am incredibly grateful that this model is open source. The speed will also increase drastically with optimisations. This is just an incredible day. What a blessing. More to come!

Kyle Hessling

24,917 görüntüleme • 28 gün önce

Dario Amodei just dismantled the biggest myth in the AI industry. Open source AI isn’t free. It never was. Amodei: “It’s not free. You have to run it on inference and someone has to make it fast on inference.” For decades, open source meant something real. It meant a teenager in a basement could download the same tools as a Fortune 500 company. Could read the code. Could modify it. Could build something that competed with the giants. That was genuine democratization. That actually happened. AI is different. Fundamentally. Physically. In ways the ideology hasn’t caught up to yet. Downloading the weights is the easy part. The part that actually costs something is turning the weights into a running system. Into responses. Into intelligence operating in real time at scale. That requires compute. Power. Infrastructure. The kind measured in billions of dollars and years of construction. Amodei: “These are big models. They’re hard to do inference on. Ultimately you have to host it on the cloud. The people who host it on the cloud do inference.” The open source debate was never about who owns the model. It was always about who owns the cloud. And Amodei goes further. When a competitor drops a new open model, he doesn’t ask whether it’s open or closed. He doesn’t care about the licensing. He doesn’t engage the ideology. Amodei: “I don’t think it mattered that DeepSeek is open source. I think I ask, is it a good model? Is it better than us at the things that matter? That’s the only thing that I care about.” That’s the ruthless clarity of someone actually trying to win. While the media debates licensing frameworks, Amodei is asking one question. Is it better. Everything else is a distraction. Amodei: “I don’t think open source works the same way in AI that it has worked in other areas. Here we can’t see inside the model.” This isn’t Linux. You can’t read it. You can’t fork it. You can’t understand it the way generations of developers understood the tools they inherited. You can download it. And then you need a data center to run it. The teenager in the basement who was supposed to be empowered by this revolution needs a billion dollars of infrastructure before the empowerment starts. The era of the basement coder rewriting civilization on a laptop is over. The future belongs to whoever commands the compute, owns the power grid, and can actually turn the intelligence on. Open weights without infrastructure isn’t democratization. It’s a promise the physics of the universe won’t let us keep.
1:40

Dario Amodei just dismantled the biggest myth in the AI industry. Open source AI isn’t free. It never was. Amodei: “It’s not free. You have to run it on inference and someone has to make it fast on inference.” For decades, open source meant something real. It meant a teenager in a basement could download the same tools as a Fortune 500 company. Could read the code. Could modify it. Could build something that competed with the giants. That was genuine democratization. That actually happened. AI is different. Fundamentally. Physically. In ways the ideology hasn’t caught up to yet. Downloading the weights is the easy part. The part that actually costs something is turning the weights into a running system. Into responses. Into intelligence operating in real time at scale. That requires compute. Power. Infrastructure. The kind measured in billions of dollars and years of construction. Amodei: “These are big models. They’re hard to do inference on. Ultimately you have to host it on the cloud. The people who host it on the cloud do inference.” The open source debate was never about who owns the model. It was always about who owns the cloud. And Amodei goes further. When a competitor drops a new open model, he doesn’t ask whether it’s open or closed. He doesn’t care about the licensing. He doesn’t engage the ideology. Amodei: “I don’t think it mattered that DeepSeek is open source. I think I ask, is it a good model? Is it better than us at the things that matter? That’s the only thing that I care about.” That’s the ruthless clarity of someone actually trying to win. While the media debates licensing frameworks, Amodei is asking one question. Is it better. Everything else is a distraction. Amodei: “I don’t think open source works the same way in AI that it has worked in other areas. Here we can’t see inside the model.” This isn’t Linux. You can’t read it. You can’t fork it. You can’t understand it the way generations of developers understood the tools they inherited. You can download it. And then you need a data center to run it. The teenager in the basement who was supposed to be empowered by this revolution needs a billion dollars of infrastructure before the empowerment starts. The era of the basement coder rewriting civilization on a laptop is over. The future belongs to whoever commands the compute, owns the power grid, and can actually turn the intelligence on. Open weights without infrastructure isn’t democratization. It’s a promise the physics of the universe won’t let us keep.

Dustin

687,842 görüntüleme • 6 ay önce

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.
0:26

Announcing the Open Model Series: a new series on open source AI models, with merve of Hugging Face. Hugging Face is the open model hub: it hosts model weights, datasets, evals, & demo apps, and maintains the open-source libraries (like transformers) used to run and train them. I think of it like GitHub, but for AI models. You can download open models and run them on your own machines. In this series we'll cover what open source and open weight models are, how they work, how to run them yourself, and why they're important. By the way - this video is edited entirely with Claude! I can make a tutorial on how I did it. If there's a question you want answered about open models, leave it in the comments. We will read them all. Episode 1 is out soon.

Harper Carroll

30,392 görüntüleme • 1 ay önce

【Live2D Showcase】 Polaire Vanilla 2.0 It's finally done! I had a lot of fun, and was able to experiment and learn a lot while working on this model. I'll be uploading a video explaining the model in more detail, so please look forward to it ✨ 🎨: CureYunny 🛠️: JakobHallows
1:32

【Live2D Showcase】 Polaire Vanilla 2.0 It's finally done! I had a lot of fun, and was able to experiment and learn a lot while working on this model. I'll be uploading a video explaining the model in more detail, so please look forward to it ✨ 🎨: CureYunny 🛠️: JakobHallows

Jakob Hallows 😈🦇 ジェイコブ・ハローズ

52,625 görüntüleme • 1 yıl önce

Jason Calacanis: Anthropic, OpenAI and others are trying to kill OpenClaw Why? Because an open source agent platform is an existential threat to frontier model companies. @jason: “ People are going to say I'm a conspiracy theorist, but the number one goal, I believe, in the large language model, frontier model space, is to kill (OpenClaw). This is a giant movement to stop it, because this is the equivalent of having an open source Android-like player in the market, and that could be incredibly disruptive. Because, I believe, open source is going to win the day on the large language models and take 90% of the token usage, and I think the entire frontier model space could be undercut by open source. And I think they realize that SLMs, the smaller language models that are verticalized now, that will run on desktops and laptops and are even starting to run on the top ones, that is their biggest competitive threat, and I hope it happens.”
0:41

Jason Calacanis: Anthropic, OpenAI and others are trying to kill OpenClaw Why? Because an open source agent platform is an existential threat to frontier model companies. @jason: “ People are going to say I'm a conspiracy theorist, but the number one goal, I believe, in the large language model, frontier model space, is to kill (OpenClaw). This is a giant movement to stop it, because this is the equivalent of having an open source Android-like player in the market, and that could be incredibly disruptive. Because, I believe, open source is going to win the day on the large language models and take 90% of the token usage, and I think the entire frontier model space could be undercut by open source. And I think they realize that SLMs, the smaller language models that are verticalized now, that will run on desktops and laptops and are even starting to run on the top ones, that is their biggest competitive threat, and I hope it happens.”

The All-In Podcast

64,677 görüntüleme • 5 ay önce

.Marques Brownlee during his Tesla Model Y Performance review: “The thing about testing one and actually driving a Tesla for a while, you start to remember all of the ways that Tesla is still ahead of the competition. I finally got my hands on the Tesla Model Y Performance— driving it has reminded me of the things that Tesla is still the best at.”
0:34

.Marques Brownlee during his Tesla Model Y Performance review: “The thing about testing one and actually driving a Tesla for a while, you start to remember all of the ways that Tesla is still ahead of the competition. I finally got my hands on the Tesla Model Y Performance— driving it has reminded me of the things that Tesla is still the best at.”

Nic Cruz Patane

320,347 görüntüleme • 7 ay önce

The MiniMax M2 model is mind-blowing! It's open-source. It outperforms Gemini 2.5, Claude 4.1, and Qwen3 across coding and tool-use benchmarks. Right now, it's one of the world's top 5 models in intelligence! And here is the best part: Claude is one of the best models you can use today, and MiniMax M2 costs only 8% of that! It's smaller, faster, and cheaper. Extremely efficient at using tokens. Minimax M2's biggest strength: High agentic capabilities. The model can plan and execute complex multi-tool workflows. It's reliable and very robust at executing long-horizon tool chains. In summary: • Low latency • Very cheap • Excels at agentic tasks • Open-source The model currently powers the MiniMax Agent and is available for a free global trial. You can access MiniMax M2's API here: To access the agent: And here is the MiniMax website: Thanks to the MiniMax team for showing me the ropes and partnering with me on this post.
6:01

The MiniMax M2 model is mind-blowing! It's open-source. It outperforms Gemini 2.5, Claude 4.1, and Qwen3 across coding and tool-use benchmarks. Right now, it's one of the world's top 5 models in intelligence! And here is the best part: Claude is one of the best models you can use today, and MiniMax M2 costs only 8% of that! It's smaller, faster, and cheaper. Extremely efficient at using tokens. Minimax M2's biggest strength: High agentic capabilities. The model can plan and execute complex multi-tool workflows. It's reliable and very robust at executing long-horizon tool chains. In summary: • Low latency • Very cheap • Excels at agentic tasks • Open-source The model currently powers the MiniMax Agent and is available for a free global trial. You can access MiniMax M2's API here: To access the agent: And here is the MiniMax website: Thanks to the MiniMax team for showing me the ropes and partnering with me on this post.

Santiago

91,197 görüntüleme • 10 ay önce

Agent Categories and Costs are now in the Agent Leaderboard! Two questions guide model selection: task-specific performance and cost to completion. Based on analysis of 1.7M+ real-world Agent Mode sessions, the leaderboard now lets users: - Filter the leaderboard by Code, Work, and Chat to compare model performance by task type - View real cost-per-task for each model, including the Pareto frontier of the most cost-efficient options at each performance level
0:38

Agent Categories and Costs are now in the Agent Leaderboard! Two questions guide model selection: task-specific performance and cost to completion. Based on analysis of 1.7M+ real-world Agent Mode sessions, the leaderboard now lets users: - Filter the leaderboard by Code, Work, and Chat to compare model performance by task type - View real cost-per-task for each model, including the Pareto frontier of the most cost-efficient options at each performance level

Arena.ai

29,227 görüntüleme • 24 gün önce

This is the craziest model you'll see this week! This model powers Figma. It's also behind Wayfair and a bunch of major e-commerce retailers. This model is a PIXEL-PERFECT image editing model. Literally, "pixel perfect". The model is capable of modifying one image while keeping everything else untouched. It's a model designed, developed, and deployed by the research team at Jasper. This beats OpenAI models, Nano Banana, and every single general image model I've seen so far. It's up to 100x less expensive to run, and up to 10X faster. I recorded a quick video to show you how impressive this is. There's a ton of research behind this model. I'm adding two links below so you can read about how this model works, specifically about "Latent Bridge Matching" and "Flash Diffusion".
4:29

This is the craziest model you'll see this week! This model powers Figma. It's also behind Wayfair and a bunch of major e-commerce retailers. This model is a PIXEL-PERFECT image editing model. Literally, "pixel perfect". The model is capable of modifying one image while keeping everything else untouched. It's a model designed, developed, and deployed by the research team at Jasper. This beats OpenAI models, Nano Banana, and every single general image model I've seen so far. It's up to 100x less expensive to run, and up to 10X faster. I recorded a quick video to show you how impressive this is. There's a ton of research behind this model. I'm adding two links below so you can read about how this model works, specifically about "Latent Bridge Matching" and "Flash Diffusion".

Santiago

25,861 görüntüleme • 10 ay önce

Introducing MIRA. A playable, multiplayer world model. A dream of Rocket League. Trained on 10k hours of data collected with publicly available bots, MIRA learns the dynamics of a four-player game. The model runs in real time at 20 fps, based on the keys you and the other players press. Built by General Intuition and kyutai, in collaboration with Epic Games. Not used to develop Rocket League. ▶️ Play the demo, read the technical report, and explore the open-source code at At ICML? Find us at Booth 111 to try it yourself and dig into the results with the team.
0:23

Introducing MIRA. A playable, multiplayer world model. A dream of Rocket League. Trained on 10k hours of data collected with publicly available bots, MIRA learns the dynamics of a four-player game. The model runs in real time at 20 fps, based on the keys you and the other players press. Built by General Intuition and kyutai, in collaboration with Epic Games. Not used to develop Rocket League. ▶️ Play the demo, read the technical report, and explore the open-source code at At ICML? Find us at Booth 111 to try it yourself and dig into the results with the team.

General Intuition

503,308 görüntüleme • 2 ay önce

Exciting News from Open-Sora! 🚀 They've just made the ENTIRE suite of their video-generation model open source! Dive into the world of cutting-edge AI with access to model weights, comprehensive training source code, and detailed architecture insights. Start building your dream video-generation model today! Check it out 👉
0:28

Exciting News from Open-Sora! 🚀 They've just made the ENTIRE suite of their video-generation model open source! Dive into the world of cutting-edge AI with access to model weights, comprehensive training source code, and detailed architecture insights. Start building your dream video-generation model today! Check it out 👉

Yang You

245,742 görüntüleme • 2 yıl önce

A nice aspect of our new Gemini 3.6 Flash model is that it is much more token efficient than our 3.5 Flash model. Here's a side-by-side demonstration of that. Nice work by everyone who worked on this model and release!
1:11

A nice aspect of our new Gemini 3.6 Flash model is that it is much more token efficient than our 3.5 Flash model. Here's a side-by-side demonstration of that. Nice work by everyone who worked on this model and release!

Jeff Dean

210,485 görüntüleme • 1 ay önce

I made a FURRY Vtuber model for Shylily! I was in charge of both the model art and rigging! Bet you wanna know what it's capable of ~ #ShylilyArt
0:30

I made a FURRY Vtuber model for Shylily! I was in charge of both the model art and rigging! Bet you wanna know what it's capable of ~ #ShylilyArt

WFN ウィスパー🐮@VeXpo

384,632 görüntüleme • 2 ay önce