lmxyy

3 ms·

1.
VC-Attention: Faster Low-Bit Attention Without Retraining(nunchux.ai)
3 points by lmxyy 1mo ago|0 comments
2.
Radial Attention: O(nlogn) Attention for Long Video Generation with 2-4× Speedup(hanlab.mit.edu)
6 points by lmxyy 1y ago|1 comments
3.
by lmxyy 1y ago
Introduce Radial Attention — a static sparse attention mechanism with O(nlogn) complexity for long video generation! Here are some key features: * Plug-and-play: works with pretrained models like Wan, HunyuanVideo, Mochi * Speeds up both tr
4.
by lmxyy 2y ago
Thanks for pointing this out. I've fixed the prompt. Both FLUX and PixArt use T5 for the text encoder, which has limited capability. Our quantization method can preserve the image quality and contents of the original 16-bit ones well.
5.
by lmxyy 2y ago
It has already released as in https://github.com/mit-han-lab/nunchaku?tab=readme-ov-file#c... .
6.
by lmxyy 2y ago
I think so. There are already some techniques called rotation, which have similar effects. But it will incur additional overheads in diffusion models.
7.
by lmxyy 2y ago
FLUX-schnell is only 800ms on RTX 5090.
8.
SVDQuant+NVFP4: 4× Smaller, 3× Faster FLUX with 16-bit Quality on Blackwell GPUs(hanlab.mit.edu)
52 points by lmxyy 2y ago|10 comments
9.
by lmxyy 2y ago
SVDQuant supports NVFP4 on NVIDIA Blackwell GPUs with 3× speedup over BF16 and better image quality than INT4. Try our interactive demo below or at https://svdquant.mit.edu/ ! Our code is all available at https://
10.
RTX 5090 Workstation Configuration Journey(hanlab.mit.edu)
5 points by lmxyy 2y ago|1 comments
11.
by lmxyy 2y ago
With the arrival of the RTX 5090, we built a high-performance workstation to maximize its AI computing potential. In this blog post, we share our experience—from overcoming setup challenges to testing its performance.
12.
SVDQuant: 4-Bit Quantization Powers 12B Flux on a 16GB 4090 GPU with 3x Speedup(hanlab.mit.edu)
179 points by lmxyy 2y ago|65 comments