Loading video...

Video Failed to Load

Go Home

Qwen3.6-27B running 100% on WebGPU. Not the best speed but still 😁

ngxson's profile picture

Xuan-Son Nguyen

6,868 subscribers

53,142 views • 3 months ago •via X (Twitter)

LIVE

Anya Rossi• Live Now

Private livecam show

0 Comments

No comments available

Comments from the original post will appear here

Related Videos

Bonsai 27B running locally on an iPhone in Atomic Chat! Bonsai is the first 27B-class model that fits on a phone. PrismML built it on Qwen3.6 27B with 1-bit weights. It takes 3.9GB instead of 54GB and keeps ~90% of the benchmark scores. Available now on iPhone and Android
0:11

Bonsai 27B running locally on an iPhone in Atomic Chat! Bonsai is the first 27B-class model that fits on a phone. PrismML built it on Qwen3.6 27B with 1-bit weights. It takes 3.9GB instead of 54GB and keeps ~90% of the benchmark scores. Available now on iPhone and Android

atomic.chat

55,237 views • 1 month ago

🥊 Qwen3.6 35B A3B vs Qwen3.6 27B Made them fight on the same prompt 🌊 Ocean Waves, a canvas challenge 35B took 39.7s (at 142tok/s) 27B took 111.6s (at 50tok/s) Both models are really good. The fact that the 35B produced such a strong result in under 40 seconds is seriously impressive. But when you look at 27B's output... it's actually so much better Added clouds, beautiful foam particles and splashing effects, a perfectly straight horizon, more realistic blinking stars, and a nice sun reflection on the water 🏆 Qwen3.6 27B slower but the result slaps Absolutely amazing that both models run smoothly on 16GB VRAM. Used Unsloth AI GGUFs here, Q3_K_S quant to make it fit in the hardware 🔥
0:16

🥊 Qwen3.6 35B A3B vs Qwen3.6 27B Made them fight on the same prompt 🌊 Ocean Waves, a canvas challenge 35B took 39.7s (at 142tok/s) 27B took 111.6s (at 50tok/s) Both models are really good. The fact that the 35B produced such a strong result in under 40 seconds is seriously impressive. But when you look at 27B's output... it's actually so much better Added clouds, beautiful foam particles and splashing effects, a perfectly straight horizon, more realistic blinking stars, and a nice sun reflection on the water 🏆 Qwen3.6 27B slower but the result slaps Absolutely amazing that both models run smoothly on 16GB VRAM. Used Unsloth AI GGUFs here, Q3_K_S quant to make it fit in the hardware 🔥

left curve dev

61,283 views • 4 months ago

Compared Qwen3.6 35B and 27B in the same conditions with Google TurboQuant Device: MacBook Pro M5Max 64GB RAM Outputs characteristics: Qwen3.6 35B: 6672 tokens, 2m 10s, 65 tok/s Qwen3.6 27B: 7344 tokens, 5m 22s, 24 tok/s Conclusion: Both models were asked to draw waves using HTML, 35B responded quickly but the result feels weak and messy, while 27B took more time and delivered a much cleaner and more consistent result, because it is built for thinking and planning, so it works better on tasks that need structure, overall 27B is a better choice for tasks where planning matters, while 35B is more suitable for everyday use when you just need a fast response
0:11

Compared Qwen3.6 35B and 27B in the same conditions with Google TurboQuant Device: MacBook Pro M5Max 64GB RAM Outputs characteristics: Qwen3.6 35B: 6672 tokens, 2m 10s, 65 tok/s Qwen3.6 27B: 7344 tokens, 5m 22s, 24 tok/s Conclusion: Both models were asked to draw waves using HTML, 35B responded quickly but the result feels weak and messy, while 27B took more time and delivered a much cleaner and more consistent result, because it is built for thinking and planning, so it works better on tasks that need structure, overall 27B is a better choice for tasks where planning matters, while 35B is more suitable for everyday use when you just need a fast response

atomic.chat

55,540 views • 4 months ago

The new PrismML’s Bonsai 27B model running on iPhone 17 Pro Bonsai 27B is the first 27B-class model to run on a phone and it uses less than 5GB of memory Pushing the limits on what can run a phone
0:14

The new PrismML’s Bonsai 27B model running on iPhone 17 Pro Bonsai 27B is the first 27B-class model to run on a phone and it uses less than 5GB of memory Pushing the limits on what can run a phone

Adrien Grondin

67,095 views • 2 months ago

Still running. Still hungry. Still the best ride you'll ever have. 🤍
0:11

Still running. Still hungry. Still the best ride you'll ever have. 🤍

nono ♡

18,003 views • 3 months ago

There is a speed limit on the road, but not offroad
0:16

There is a speed limit on the road, but not offroad

Criminal Tube

53,038 views • 1 year ago

The speed of this merganser running on water
0:12

The speed of this merganser running on water

Science girl

7,728,604 views • 3 years ago

The speed of a merganser running on water.
0:12

The speed of a merganser running on water.

Wonder of Science

795,780 views • 2 years ago

The speed of this merganser running on water
0:12

The speed of this merganser running on water

Nature & Animals🌴

15,595 views • 27 days ago

WebGPU ShaderToy in Flutter. Not possible to build this on mainline Flutter. Trivial in my WebGPU / Graphite based fork.
0:10

WebGPU ShaderToy in Flutter. Not possible to build this on mainline Flutter. Trivial in my WebGPU / Graphite based fork.

Jesse Ezell

10,091 views • 2 months ago

Falken's Porsche is in the fence! The best 911 was running P5, but not anymore... 📺 #IGTC | #24hNBR
0:16

Falken's Porsche is in the fence! The best 911 was running P5, but not anymore... 📺 #IGTC | #24hNBR

Intercontinental GT Challenge

53,286 views • 4 months ago

“He’s better than a Justin Fields type running quarterback, but he’s still not a Baker Mayfield throwing the ball.” Put some respect on Lamar’s name.
0:17

“He’s better than a Justin Fields type running quarterback, but he’s still not a Baker Mayfield throwing the ball.” Put some respect on Lamar’s name.

The Ravens Realm

204,996 views • 1 year ago

WOW! 🤯 Language models are becoming smaller and more capable than ever! Here's SmolLM2 running 100% locally in-browser w/ WebGPU on a 6-year-old GPU. Look at that speed! ⚡️😍 Powered by 🤗 Transformers.js and ONNX Runtime Web! How many tokens/second do you get? Let me know! 👇
0:13

WOW! 🤯 Language models are becoming smaller and more capable than ever! Here's SmolLM2 running 100% locally in-browser w/ WebGPU on a 6-year-old GPU. Look at that speed! ⚡️😍 Powered by 🤗 Transformers.js and ONNX Runtime Web! How many tokens/second do you get? Let me know! 👇

Xenova

12,557 views • 1 year ago

Day-0 Qwen3.8-27B vs Qwen3.6-27B: the voxel pagoda test. Same prompt, one attempt each, single DGX Spark, both NVFP4. 3.8: 75,291 tok · 52m 56s · 23.7 tok/s (vLLM, native MTP) 3.6: 12,117 tok · 5m 13s · 38.8 tok/s (llama.cpp, DFlash K10) 3.8 thought for 160K characters before writing a line of code.
0:15

Day-0 Qwen3.8-27B vs Qwen3.6-27B: the voxel pagoda test. Same prompt, one attempt each, single DGX Spark, both NVFP4. 3.8: 75,291 tok · 52m 56s · 23.7 tok/s (vLLM, native MTP) 3.6: 12,117 tok · 5m 13s · 38.8 tok/s (llama.cpp, DFlash K10) 3.8 thought for 160K characters before writing a line of code.

Wësche

97,988 views • 1 month ago

Stable Diffusion 100% in the browser with running at < 1s per image ⚡️ Thanks to SD Turbo by Axel Sauer & al Stability AI and diffusers.js on webGPU by Arthur Islamov Demo: More info in ↓🧵
0:13

Stable Diffusion 100% in the browser with running at < 1s per image ⚡️ Thanks to SD Turbo by Axel Sauer & al Stability AI and diffusers.js on webGPU by Arthur Islamov Demo: More info in ↓🧵

Cyril Diagne

96,635 views • 2 years ago

Qwen3.5 27B vs Gemma4 31B | Canvas Creativity Test Why HTML Canvas? Two reasons: 1. It's unforgiving, one small mistake and the whole thing breaks 2. We kept prompts short to test real creativity, not instruction following 4 rounds: - Analog Clock - Hyperspace Tunnel - Growing Tree - Black Hole Both nailed the clock, but the other three is where it gets interesting. Looking forward to Qwen3.6 open-weight release!
0:17

Qwen3.5 27B vs Gemma4 31B | Canvas Creativity Test Why HTML Canvas? Two reasons: 1. It's unforgiving, one small mistake and the whole thing breaks 2. We kept prompts short to test real creativity, not instruction following 4 rounds: - Analog Clock - Hyperspace Tunnel - Growing Tree - Black Hole Both nailed the clock, but the other three is where it gets interesting. Looking forward to Qwen3.6 open-weight release!

stevibe

170,768 views • 5 months ago

Not sure what the speed limit is on this road but 50-ish seems good
0:13

Not sure what the speed limit is on this road but 50-ish seems good

Ronnie Adkins

15,899 views • 1 month ago

290MB = useful LLM (running directly in your browser via WebGPU) 🤯 Bonsai 1-bit is beyond my comprehension... try the demo on hugging face ⬇️
0:10

290MB = useful LLM (running directly in your browser via WebGPU) 🤯 Bonsai 1-bit is beyond my comprehension... try the demo on hugging face ⬇️

Victor M

63,244 views • 5 months ago

not to be biased but this is still the best event ui genshin's ever done whoever did this needs to be kissed on the mouth
0:13

not to be biased but this is still the best event ui genshin's ever done whoever did this needs to be kissed on the mouth

kao the genshin pilot 🍀

304,707 views • 1 year ago