Загрузка видео...

Не удалось загрузить видео

На главную

Still didn't attempt any quality testing, but let's say that GPT 5.6 Sol was able almost unassisted to write this Laguna S2.1 inference implementation following the other two models in DwarfStar, and it is very fast, 50 t/s generation, very fast prefill as well.

antirez's profile picture

antirez

78,104 subscribers

43,538 просмотров • 29 дней назад •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

Комментарии: 0

Нет доступных комментариев

Здесь появятся комментарии из оригинального поста

Похожие видео

I didn't expect DeepSeek v4 PRO (not Flash) to run well on the Mac Studio M3 Ultra with 512GB of RAM. This is 2 bit quantized with the same DwarfStar recipe used for Flash. 433GB GGUF file. 130 t/s prefill, 13 t/s generation. Prefill in the video is low because small prompt.
1:24

I didn't expect DeepSeek v4 PRO (not Flash) to run well on the Mac Studio M3 Ultra with 512GB of RAM. This is 2 bit quantized with the same DwarfStar recipe used for Flash. 433GB GGUF file. 130 t/s prefill, 13 t/s generation. Prefill in the video is low because small prompt.

antirez

171,478 просмотров • 3 месяцев назад

For the DGX Spark owners. This is what you get with DS4 in your hardware. I want to post this to show how with fast prefill and not very fast generation, the system remains absolutely fine to use.
1:54

For the DGX Spark owners. This is what you get with DS4 in your hardware. I want to post this to show how with fast prefill and not very fast generation, the system remains absolutely fine to use.

antirez

40,085 просмотров • 2 месяцев назад

well, i connected GPT-5.6 Sol to Blender and... let's just say the world of 3D graphics (at least via Blender's MCP) is safe for now... but still... this whole scene was generated by GPT-5.6 with very little human intervention feels like *EARLY* pixar, like too early lol
1:30

well, i connected GPT-5.6 Sol to Blender and... let's just say the world of 3D graphics (at least via Blender's MCP) is safe for now... but still... this whole scene was generated by GPT-5.6 with very little human intervention feels like *EARLY* pixar, like too early lol

Gavin Purcell

62,349 просмотров • 1 месяц назад

DeepSeek v4 PRO running in DwarfStar on a DGX Station. Can go faster, already peaks 50 t/s without DSpark: dynamic VRAM/RAM allocation of experts based on past statistics, 4k t/s prefill on big chunks. VRAM/RAM layer-major streaming. Mixed RAM/VRAM inference requires care.
0:27

DeepSeek v4 PRO running in DwarfStar on a DGX Station. Can go faster, already peaks 50 t/s without DSpark: dynamic VRAM/RAM allocation of experts based on past statistics, 4k t/s prefill on big chunks. VRAM/RAM layer-major streaming. Mixed RAM/VRAM inference requires care.

antirez

24,823 просмотров • 3 дней назад

meta just dropped a bomb. Muse Spark 1.1 is 86% cheaper than GPT-5.6 Sol and 91% cheaper than Fable 5 on output tokens ($4.25 vs $30 vs $50 per M) I thought the model was going to be horrendous, but it ended up being insanely fast, and very capable. front end design is VERY strong. Agentic capabilities seem great as it was able to spawn and even orchestrate other agents inside my app cnvs using the mcp and cli tool with zero instruction. we have soo many insanely good models right now.
0:56

meta just dropped a bomb. Muse Spark 1.1 is 86% cheaper than GPT-5.6 Sol and 91% cheaper than Fable 5 on output tokens ($4.25 vs $30 vs $50 per M) I thought the model was going to be horrendous, but it ended up being insanely fast, and very capable. front end design is VERY strong. Agentic capabilities seem great as it was able to spawn and even orchestrate other agents inside my app cnvs using the mcp and cli tool with zero instruction. we have soo many insanely good models right now.

Max Blade

88,457 просмотров • 1 месяц назад

DGX Station: this is just the start of what it is possible to do (and why I claim DwarfStar could be *the* Station inference engine). DeepSeek v4 PRO Q2 with routed experts split among VRAM / RAM with kernels optimized for this peculiar setup. 45 t/s but can go faster.
0:53

DGX Station: this is just the start of what it is possible to do (and why I claim DwarfStar could be *the* Station inference engine). DeepSeek v4 PRO Q2 with routed experts split among VRAM / RAM with kernels optimized for this peculiar setup. 45 t/s but can go faster.

antirez

41,545 просмотров • 4 дней назад

Crazy prefill speed of DwarfStar running DeepSeek v4 PRO in the DGX Station (private branch for now) using techniques that leverage the peculiar RAM + VRAM mixed setup of the Station. 3000 t/s but even more can be reached.
0:23

Crazy prefill speed of DwarfStar running DeepSeek v4 PRO in the DGX Station (private branch for now) using techniques that leverage the peculiar RAM + VRAM mixed setup of the Station. 3000 t/s but even more can be reached.

antirez

60,185 просмотров • 2 дней назад

DS4 running on DGX Spark (GB10 / CUDA), private branch for now. 12 tokens/sec, the memory bandwidth is limited in this system, at 270GB/sec. But prefill is ways more alighed to M3 Max at ~200 t/s. I'll release when more mature, but it is almost sure that it will get merged.
1:34

DS4 running on DGX Spark (GB10 / CUDA), private branch for now. 12 tokens/sec, the memory bandwidth is limited in this system, at 270GB/sec. But prefill is ways more alighed to M3 Max at ~200 t/s. I'll release when more mature, but it is almost sure that it will get merged.

antirez

84,127 просмотров • 3 месяцев назад

Two frontier models. Two very different demos. (GPT-5.6 Sol and Claude Fable 5) GPT-5.6 Sol Ultra was pushed in a mathematical direction and generated a Minecraft-style clone in Lean. Yes, that Lean. Meanwhile, Claude Fable 5 took a game development route and built a Minecraft mod featuring a Dark Souls-inspired enemy with multiple attack patterns, combos, and combat logic. They're solving completely different problems, but both are great examples of how capable today's models have become. The pace of progress is remarkable.
0:49

Two frontier models. Two very different demos. (GPT-5.6 Sol and Claude Fable 5) GPT-5.6 Sol Ultra was pushed in a mathematical direction and generated a Minecraft-style clone in Lean. Yes, that Lean. Meanwhile, Claude Fable 5 took a game development route and built a Minecraft mod featuring a Dark Souls-inspired enemy with multiple attack patterns, combos, and combat logic. They're solving completely different problems, but both are great examples of how capable today's models have become. The pace of progress is remarkable.

FHILY👑

19,336 просмотров • 1 месяц назад

GPT-5.6 vs GPT-5.5 on my custom spaceship prompt. I gave both models the exact same custom prompt. This is also the same prompt I previously gave to Fable 5. For context, GPT-5.6 Pro worked for 87 minutes, while GPT-5.5 Extra High worked for 34 minutes and 42 seconds. As I’ve said before, based on great authority GPT-5.6 will be an incremental/soldi improvement over GPT-5.5, not a “Fable killer.” My rough expectation has been that it would trade blows with Fable 5 on some benchmarks, maybe win around half depending on the category, but not clearly surpass it overall. And again fable five will have bigger model smell, but this was expected. After testing this coding output, that view feels pretty accurate. GPT-5.6 is clearly better than GPT-5.5 in several visual areas. The lighting, shading, chairs, object details, and exterior of the spaceship looked noticeably stronger. The scene was also easier to test. I do want to give GPT-5.5 credit though. It built out the rooms much much better and the planets looked better than GPT-5.6’s. It was also interesting that both GPT-5.5 and GPT-5.6 produced better-looking planets than Fable 5 in this specific test. The downside with GPT-5.5 was stability. The game was much glitchier and harder to test compared to GPT-5.6. But when it comes to the core of the demo, which is the spaceship itself, Fable 5 still beat both models pretty comfortably. GPT-5.6 is impressive, but from this test, it looks exactly like what I expected which was a meaningful incremental improvement over GPT-5.5, at least for indie game demos, but not something that replaces Fable 5. In collaboration with Chetaslua
0:42

GPT-5.6 vs GPT-5.5 on my custom spaceship prompt. I gave both models the exact same custom prompt. This is also the same prompt I previously gave to Fable 5. For context, GPT-5.6 Pro worked for 87 minutes, while GPT-5.5 Extra High worked for 34 minutes and 42 seconds. As I’ve said before, based on great authority GPT-5.6 will be an incremental/soldi improvement over GPT-5.5, not a “Fable killer.” My rough expectation has been that it would trade blows with Fable 5 on some benchmarks, maybe win around half depending on the category, but not clearly surpass it overall. And again fable five will have bigger model smell, but this was expected. After testing this coding output, that view feels pretty accurate. GPT-5.6 is clearly better than GPT-5.5 in several visual areas. The lighting, shading, chairs, object details, and exterior of the spaceship looked noticeably stronger. The scene was also easier to test. I do want to give GPT-5.5 credit though. It built out the rooms much much better and the planets looked better than GPT-5.6’s. It was also interesting that both GPT-5.5 and GPT-5.6 produced better-looking planets than Fable 5 in this specific test. The downside with GPT-5.5 was stability. The game was much glitchier and harder to test compared to GPT-5.6. But when it comes to the core of the demo, which is the spaceship itself, Fable 5 still beat both models pretty comfortably. GPT-5.6 is impressive, but from this test, it looks exactly like what I expected which was a meaningful incremental improvement over GPT-5.5, at least for indie game demos, but not something that replaces Fable 5. In collaboration with Chetaslua

Chris

250,405 просмотров • 2 месяцев назад

First kinda working implementation of GLM 5.2 in DwarfStar. Will take some time to be good enough, but it is a promising start. 433 GB GGUF file. M3 Ultra 512GB.
0:36

First kinda working implementation of GLM 5.2 in DwarfStar. Will take some time to be good enough, but it is a promising start. 433 GB GGUF file. M3 Ultra 512GB.

antirez

69,508 просмотров • 2 месяцев назад

Took the good work of the communtiy of DwarfStar and consolidating the Strix Halo support. It looks very good. More QA in the next days and the final merge soon.
1:20

Took the good work of the communtiy of DwarfStar and consolidating the Strix Halo support. It looks very good. More QA in the next days and the final merge soon.

antirez

17,303 просмотров • 2 месяцев назад

Trump: Drain the swamp, and that's what we're doing. When I said it, it sounded very easy, and it was gonna happen real fast. I didn't know the swamp was this dirty and this deep. I didn't know about the swamp creatures being as bad as they are. They're bad. But nobody can believe what we've been able to do.
0:27

Trump: Drain the swamp, and that's what we're doing. When I said it, it sounded very easy, and it was gonna happen real fast. I didn't know the swamp was this dirty and this deep. I didn't know about the swamp creatures being as bad as they are. They're bad. But nobody can believe what we've been able to do.

FactPost

65,574 просмотров • 2 месяцев назад

Nvidia $NVDA CEO Jensen Huang was asked about a recent story that said he warned that China will beat the US in the AI race “That’s not what I said. What I said was China has very good AI technology. They have many AI researchers, in fact 50% of the world’s AI researchers are in China. And they develop very good AI technology. In fact, the most popular AI models in the world today, open-source models, are from China. So, they are moving very, very fast. The United States has to continue to move incredibly fast. And otherwise, otherwise – the world is very competitive, so we have to run fast.” (H/T Dan Nystedt)
0:42

Nvidia $NVDA CEO Jensen Huang was asked about a recent story that said he warned that China will beat the US in the AI race “That’s not what I said. What I said was China has very good AI technology. They have many AI researchers, in fact 50% of the world’s AI researchers are in China. And they develop very good AI technology. In fact, the most popular AI models in the world today, open-source models, are from China. So, they are moving very, very fast. The United States has to continue to move incredibly fast. And otherwise, otherwise – the world is very competitive, so we have to run fast.” (H/T Dan Nystedt)

Evan

1,946,899 просмотров • 9 месяцев назад

a really good (and cheap!) in-between is hooking up an old laptop/PC to a CRT via HDMI-to-Composite and loading it with RetroArch/Emulation-Station. Plays almost any 5th generation and below consoles with ease. 6th generation and up is a more involved but very doable as well
1:16

a really good (and cheap!) in-between is hooking up an old laptop/PC to a CRT via HDMI-to-Composite and loading it with RetroArch/Emulation-Station. Plays almost any 5th generation and below consoles with ease. 6th generation and up is a more involved but very doable as well

tippity

75,054 просмотров • 5 месяцев назад

I think it's a great myth that music stopped being ambitious in the 80's and my new 'Intrigue' compilation is an attempt to prove that creativity, experimentation and the art rock/progressive aesthetic was still very much alive and well in the decade.
0:59

I think it's a great myth that music stopped being ambitious in the 80's and my new 'Intrigue' compilation is an attempt to prove that creativity, experimentation and the art rock/progressive aesthetic was still very much alive and well in the decade.

Steven Wilson

50,809 просмотров • 3 лет назад

Multi said he didn't really write yesterday's lore for his side because it was supposed to be Katie only, but she invited him and says he is thankful for that because it was very fun as well as Quackity, and he hopes everyone had fun #qsmp2 trio so peak fr
0:35

Multi said he didn't really write yesterday's lore for his side because it was supposed to be Katie only, but she invited him and says he is thankful for that because it was very fun as well as Quackity, and he hopes everyone had fun #qsmp2 trio so peak fr

giddy

93,160 просмотров • 2 месяцев назад