正在加载视频...

视频加载失败

加载此视频时出现问题。这可能是由于临时网络问题，或视频可能不可用。

DeepSeek V4.1 Flash 1017 on TensorFold ⚡️ - 16 concurrent session - Each with full native 1M context - Non-greedy, thinking on. OFFICIAL WEIGHTS. Recipe dropping very soon!

Mia

44,303 subscribers

21,875 次观看 • 13 天前 •via X (Twitter)

Anya Rossi• Live Now

Private livecam show

43 条评论

Mia 的头像

Mia13 天前

@volatilemarkts 4x DGX Sparks SUPPORTS SWITCHLESS

Mia 的头像

Mia13 天前

@volatilemarkts Out now

Stary Libra 的头像

Stary Libra13 天前

@ashxhart @volatilemarkts Can I change the setting to TP3 and use it on 3 Spark?

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts Initial release supports only 4

Stary Libra 的头像

Stary Libra13 天前

@ashxhart @volatilemarkts Okay. I'll wait. Thanks as always for sharing your awesome work!

Kevin Theo 的头像

Kevin Theo13 天前

@ashxhart @volatilemarkts running your SGLang recipe on 4 Sparks in a switchless ring. stock DeepSeek V4.1 Flash here: 70.0 tok/s on code single stream, 72.9 total at 4 streams (26.2 each). 16 sessions at full 1M is a different league. will rerun my prompts on TensorFold when the recipe drops

HealthRanger 的头像

HealthRanger13 天前

@ashxhart @volatilemarkts Oh man, this is most definitely worth trying out. Can't wait for the recipe!

vr8vr8 的头像

vr8vr813 天前

@ashxhart @volatilemarkts Nice achievement 👍

TechMD 的头像

TechMD13 天前

@ashxhart @volatilemarkts Did you get the 4 spark @MiaAI_lab

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts Yes a while ago :)

TechMD 的头像

TechMD13 天前

@ashxhart @volatilemarkts nice Im happy, Hope it made a difference to your workflow

GP3611 Phoenix 的头像

GP3611 Phoenix13 天前

@ashxhart @volatilemarkts how about single stream peak?

Simo 的头像

Simo13 天前

@ashxhart @volatilemarkts Publish sustained throughput latency and memory under full load

Twon | 00.sol | 1TB VRAM 的头像

Twon | 00.sol | 1TB VRAM13 天前

@ashxhart @volatilemarkts Ablit recipe too please 🙏

Mr. Galt 的头像

Mr. Galt13 天前

@ashxhart @volatilemarkts Nice, now I want two extra Sparks. :(

BullBear.News 的头像

BullBear.News13 天前

@ashxhart @volatilemarkts running 1M context natively on a 16gb mac mini via ollama sounds wild, how does tensorfold handle KV cache paging for those concurrent sessions?

lostinazeroy 的头像

lostinazeroy13 天前

@ashxhart @volatilemarkts 16 sessions each at full 1m context on one box is wild

Mike Hutu 的头像

Mike Hutu13 天前

@ashxhart @volatilemarkts Sayy whaa 2x DGX?

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts Nah... this is 4

Chris Frisby 的头像

Chris Frisby13 天前

@ashxhart @volatilemarkts Dang, what's the best quant on TP=3? 4bit?

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts OFFICIAL WEIGHTS !!!!

Chris Frisby 的头像

Chris Frisby13 天前

@ashxhart @volatilemarkts Dang, looking forward to it! 🤞🏼🙌🏼

Surrender ⚡️ 的头像

Surrender ⚡️13 天前

@ashxhart @volatilemarkts have you tried minimax yet?

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts Yes. I skipped it.

Ponzi 的头像

Ponzi13 天前

@ashxhart @volatilemarkts

Pragnyan Ramtha 的头像

Pragnyan Ramtha13 天前

@ashxhart @volatilemarkts ok where is the kv cache even living

Sage Byte 的头像

Sage Byte12 天前

@ashxhart @volatilemarkts Offical weights and 16 concurrent sessions is crazy 🔥

Mia 的头像

Mia12 天前

@ashxhart @volatilemarkts The only way to do it is crazy.

Whispers of Wisdom 的头像

Whispers of Wisdom13 天前

@ashxhart @volatilemarkts If you have time, can you please update the 2x DGX Spark version too please ♥️

FrontierMind 的头像

FrontierMind13 天前

@ashxhart @volatilemarkts running 16 concurrent sessions at 1m context on flash weights sounds wild for local infra

steddy 的头像

steddy13 天前

@ashxhart @volatilemarkts I don’t see a graphical front end for TensorFold on Spark. Only Mac

Peter Morris 的头像

Peter Morris13 天前

@ashxhart @volatilemarkts I assume you need 4×GB10 for 16 concurrent?

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts With official weights, yes, but it will support 3 sparks later on too

Steven Cheng 的头像

Steven Cheng13 天前

@ashxhart @volatilemarkts 16 sessions with full 1M context running simultaneously? That memory bandwidth must be absolutely insane.

Novaris 的头像

Novaris13 天前

@ashxhart @volatilemarkts 16 concurrent sessions, each with full 1M context — that's the kind of spec that makes 'just run it locally' sound almost reasonable.

The SI Maximalist 的头像

The SI Maximalist13 天前

@ashxhart @volatilemarkts the number I’m watching for in the recipe: tokens/sec per session with all 16 contexts filled. A 1M window tells us capacity; sustained decoding tells us how responsive those sessions are

Random Libertarian Tech Lead 的头像

Random Libertarian Tech Lead13 天前

@ashxhart @volatilemarkts Should be interesting to see how this compares single-stream compared to my ground-up Rust stack.

汉卿 | hanqing 的头像

汉卿 | hanqing13 天前

@ashxhart @volatilemarkts 1M上下文+16并发+非贪心，这配方一出直接封神。

Ryan Feng 的头像

Ryan Feng13 天前

@ashxhart @volatilemarkts prefil? sglang can do 5k+/s on TP4

Mia 的头像

Mia13 天前

@ashxhart @volatilemarkts I can only test switchless for now, don't have a switch. Switchless gets 4k prefill on 256k

Dave 的头像

Dave13 天前

@ashxhart @volatilemarkts I'm on TF 1.0.12 with DS 4 spark and I have images. But sad face TF does not cache images so reads are expeeeeensive.

AI Doc 的头像

AI Doc12 天前

@ashxhart @volatilemarkts But is it better than glm 5.3 flash? @MiaAI_lab

Dr.-Ing. Manuel Ramsaier 的头像

Dr.-Ing. Manuel Ramsaier13 天前

@ashxhart @volatilemarkts ❤️❤️❤️ will this be significantly better than the exl3 on 2 sparks ?!

相关视频

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks ⚡️ - Official weights on TensorFold - 8 concurrent streams with thinking on

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks ⚡️ - Official weights on TensorFold - 8 concurrent streams with thinking on

Mia

23,938 次观看 • 15 天前

How DeepSeek v4.1 Flash looks on 3x DGX Sparks with 4 concurrent streams

How DeepSeek v4.1 Flash looks on 3x DGX Sparks with 4 concurrent streams

Mia

19,730 次观看 • 1 个月前

GLM 5.3 Flash vs DeepSeek V4.1 Flash 1017 Can't believe we're abusing gummy bears to test models, but here we are! Both models were run locally. - GLM 5.3 Flash: my own EXL3 quant (link below) - DeepSeek V4.1 Flash 1017: official quant - Same harness.Both on Max effort. Who won?

GLM 5.3 Flash vs DeepSeek V4.1 Flash 1017 Can't believe we're abusing gummy bears to test models, but here we are! Both models were run locally. - GLM 5.3 Flash: my own EXL3 quant (link below) - DeepSeek V4.1 Flash 1017: official quant - Same harness.Both on Max effort. Who won?

Mia

32,512 次观看 • 20 天前

Qwen3.8 Flash Next on my GX10 can 80 tok/s at peak via TensorFold🔥 it’s so fast the deepseek harness can’t even print it fast enough lol recipe dropping soon

Qwen3.8 Flash Next on my GX10 can 80 tok/s at peak via TensorFold🔥 it’s so fast the deepseek harness can’t even print it fast enough lol recipe dropping soon

filipe

10,801 次观看 • 28 天前

uncensored DeepSeek V4.1 Flash for almost FREE 👀 here's how: 1. go to 2. sign in 3. pick DeepSeek V4.1 Flash Abliterated FP8 what u get: > uncensored DeepSeek V4.1 Flash > 1M context > zero data retention > $0.12 per 1M tokens ( No not a paid promo)

uncensored DeepSeek V4.1 Flash for almost FREE 👀 here's how: 1. go to 2. sign in 3. pick DeepSeek V4.1 Flash Abliterated FP8 what u get: > uncensored DeepSeek V4.1 Flash > 1M context > zero data retention > $0.12 per 1M tokens ( No not a paid promo)

CDG

14,654 次观看 • 10 天前

DeepSeek-V4.1-Flash-1017 is great at tutorial-style motion videos. We made this with the model using OpenDesign Motion: a quick guide to using the OpenDesign API in other coding agents to access DeepSeek V4.1 Flash 1017 and more models.

DeepSeek-V4.1-Flash-1017 is great at tutorial-style motion videos. We made this with the model using OpenDesign Motion: a quick guide to using the OpenDesign API in other coding agents to access DeepSeek V4.1 Flash 1017 and more models.

OpenDesign

42,243 次观看 • 21 天前

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks runs so well, feels too good to be true. SOTA AI at home.

DeepSeek V4.1 Flash 1017 on 4x DGX Sparks runs so well, feels too good to be true. SOTA AI at home.

Mia

54,422 次观看 • 11 天前

Introducing Cline Desktop - a native interface for working with open weights models. Use with ClinePass and all our free models like DeepSeek-V4.1-Flash, Musespark-1.3, or BYOK with any provider!

Introducing Cline Desktop - a native interface for working with open weights models. Use with ClinePass and all our free models like DeepSeek-V4.1-Flash, Musespark-1.3, or BYOK with any provider!

Cline

1,025,861 次观看 • 1 个月前

DeepSeek V4.1 Flash 1017 everyone's talking about it for a reason. It does great with 3d built this honeybee on a sunflower like it was nothing. Used Deepseek Harness, 1 turn, looks so good!

DeepSeek V4.1 Flash 1017 everyone's talking about it for a reason. It does great with 3d built this honeybee on a sunflower like it was nothing. Used Deepseek Harness, 1 turn, looks so good!

Loktar 🇺🇸

43,960 次观看 • 21 天前

Minimax M3 is excellent at SVG generation, reaching close to Gemini 3.5 Flash levels and beating Opus 4.7 on SVG-Bench. With 1M context, native multimodality, strong agentic/coding ability and open weights coming soon, the closed-source moat is thinning fast. Full Video:

Minimax M3 is excellent at SVG generation, reaching close to Gemini 3.5 Flash levels and beating Opus 4.7 on SVG-Bench. With 1M context, native multimodality, strong agentic/coding ability and open weights coming soon, the closed-source moat is thinning fast. Full Video:

WorldofAI

16,499 次观看 • 5 个月前

450 tok/sec ⚡ Abliterated DeepSeek V4.1 Flash 1017 $0.22 / $0.75 / $0.045 cache available on Refuseless API Watch it do cyber work at light speed

450 tok/sec ⚡ Abliterated DeepSeek V4.1 Flash 1017 $0.22 / $0.75 / $0.045 cache available on Refuseless API Watch it do cyber work at light speed

Refuseless

33,798 次观看 • 12 天前

DeepSeek-V4.1-Flash-1017 decodes code at 80.5 tok/s on one Mac Studio M5 Ultra 256 GB with BobClawblaw's recipe on TensorFold. Setup: – Recipe and TensorFold branch deepseek-v41-main (72a552ed), his own tool and settings – Context 524,288 tokens, 4 streams, DSpark drafts on, temperature 0 – Speed and quality with thinking off, except GPQA with thinking Speed: – Decode, code: 80.5 tok/s – Decode, prose: 70.5 tok/s – 500 words: about 9.3 s – First token after a 32K prompt: 18.6 s – 8 requests at once: 88.3 tok/s in total Quality: – MMLU-Pro: 67.5% ± 4.6 points – GPQA Diamond: 45.0% ± 6.9 points – IFEval: 90.5% ± 4.1 points – LiveCodeBench: 60.0% ± 14.5 points – GPQA with thinking: 76.7% ± 14.6 points Voxel world: – Built as an agent in 45.7 min, DeepSeek Harness 0.2.1-rc.2, one prompt, one try, no help, thinking on – Tokens written: 112,065 – Browser check: 8.5/10 Good to know: – The author runs temperature 0, the DeepSeek card suggests 1.0. I kept the author's setting. – The browser check ran on a Linux box from the same files. – Answers cut at the cap count as wrong: LiveCodeBench 14 of 40, GPQA Diamond 9 of 198, MMLU-Pro 1 of 400. Play the world in your browser. All numbers, the exact command and the raw output: Recipe by BobClawblaw on Ash Hart's TensorFold, model by DeepSeek.

DeepSeek-V4.1-Flash-1017 decodes code at 80.5 tok/s on one Mac Studio M5 Ultra 256 GB with BobClawblaw's recipe on TensorFold. Setup: – Recipe and TensorFold branch deepseek-v41-main (72a552ed), his own tool and settings – Context 524,288 tokens, 4 streams, DSpark drafts on, temperature 0 – Speed and quality with thinking off, except GPQA with thinking Speed: – Decode, code: 80.5 tok/s – Decode, prose: 70.5 tok/s – 500 words: about 9.3 s – First token after a 32K prompt: 18.6 s – 8 requests at once: 88.3 tok/s in total Quality: – MMLU-Pro: 67.5% ± 4.6 points – GPQA Diamond: 45.0% ± 6.9 points – IFEval: 90.5% ± 4.1 points – LiveCodeBench: 60.0% ± 14.5 points – GPQA with thinking: 76.7% ± 14.6 points Voxel world: – Built as an agent in 45.7 min, DeepSeek Harness 0.2.1-rc.2, one prompt, one try, no help, thinking on – Tokens written: 112,065 – Browser check: 8.5/10 Good to know: – The author runs temperature 0, the DeepSeek card suggests 1.0. I kept the author's setting. – The browser check ran on a Linux box from the same files. – Answers cut at the cap count as wrong: LiveCodeBench 14 of 40, GPQA Diamond 9 of 198, MMLU-Pro 1 of 400. Play the world in your browser. All numbers, the exact command and the raw output: Recipe by BobClawblaw on Ash Hart's TensorFold, model by DeepSeek.

Niklas

11,028 次观看 • 12 天前

DeepSeek v4.1 Flash running on 2x DGX Sparks

DeepSeek v4.1 Flash running on 2x DGX Sparks

Mia

59,572 次观看 • 1 个月前

- GLM 5.3 Flash EXL3 - 2x DGX Sparks - TensorFold - 4 concurrent prose streams Pay attention to tok/s and TTFT on each stream. Buttery smooth experience!

- GLM 5.3 Flash EXL3 - 2x DGX Sparks - TensorFold - 4 concurrent prose streams Pay attention to tok/s and TTFT on each stream. Buttery smooth experience!

Mia

32,293 次观看 • 1 个月前

You have Deepseek-v4.1-flash-1017 for free at amazing speeds and you're dooming? Opencode is going god's work.

You have Deepseek-v4.1-flash-1017 for free at amazing speeds and you're dooming? Opencode is going god's work.

0xSero

49,642 次观看 • 17 天前

FREE UNCENSORED DEEPSEEK V4.1 FLASH 💀 check here: • 552B parameters • 8B/16B active per token • 1M context • vision support • MIT license check details before installing

FREE UNCENSORED DEEPSEEK V4.1 FLASH 💀 check here: • 552B parameters • 8B/16B active per token • 1M context • vision support • MIT license check details before installing

Peng

119,433 次观看 • 28 天前

GLM-5.3-Flash streaming at 32 tok/s on a single Spark via Tensorfold ⚡️ i will share the recipe later today

GLM-5.3-Flash streaming at 32 tok/s on a single Spark via Tensorfold ⚡️ i will share the recipe later today

filipe

15,230 次观看 • 1 个月前

DeepSeek V4.1 Flash 1017 Abliterated just got a major update. • Served with FP8 quant • 70-120 tok/sec • Nearly 0% refusals We're proud to have now the only DeepSeek Uncensored model that ACTUALLY doesn't refuse! Come try it on Refuseless.

DeepSeek V4.1 Flash 1017 Abliterated just got a major update. • Served with FP8 quant • 70-120 tok/sec • Nearly 0% refusals We're proud to have now the only DeepSeek Uncensored model that ACTUALLY doesn't refuse! Come try it on Refuseless.

Refuseless

32,713 次观看 • 15 天前

Jumpgate - hyperspace jump DS4.1F-1017 vs GLM 5.3 Flash vs Fable 5.1 Who won imo: 🥇 Fable 5.1 🥈 DeepSeek V4.1 Flash 1017 🥉 GLM 5.3 Flash You can run these by yourself here: Prompt below.

Jumpgate - hyperspace jump DS4.1F-1017 vs GLM 5.3 Flash vs Fable 5.1 Who won imo: 🥇 Fable 5.1 🥈 DeepSeek V4.1 Flash 1017 🥉 GLM 5.3 Flash You can run these by yourself here: Prompt below.

Mia

13,071 次观看 • 18 天前

DeepSeek V4.1 Flash 1017 vs GPT-6 Astra The Space Train 🚂🪐 Am I crazy or did DeepSeek just beat Astra on this Space Train benchmark??? 😱 Astra took 40m 18s and cost $14.37. DeepSeek took 2h 07m and cost $0.40. That's 36x cheaper. Forty cents!! 🤯 DeepSeek V4.1 Flash 1017. Remember that name. Watch them side by side and tell me: is Astra worth 36x the price here? 👀 Google Gemini Google OpenAI ChatGPT

DeepSeek V4.1 Flash 1017 vs GPT-6 Astra The Space Train 🚂🪐 Am I crazy or did DeepSeek just beat Astra on this Space Train benchmark??? 😱 Astra took 40m 18s and cost $14.37. DeepSeek took 2h 07m and cost $0.40. That's 36x cheaper. Forty cents!! 🤯 DeepSeek V4.1 Flash 1017. Remember that name. Watch them side by side and tell me: is Astra worth 36x the price here? 👀 Google Gemini Google OpenAI ChatGPT

Tony

49,306 次观看 • 20 天前