Ecosyste.ms: Timeline

Browse the timeline of events for every public repo on GitHub. Data updated hourly from GH Archive.

YangWang92

YangWang92 starred haroldsultan/MCTS
YangWang92 starred pasky/michi
YangWang92 pushed 1 commit to wy/profile YangWang92/VPTQ

View on GitHub

YangWang92 created a branch on YangWang92/VPTQ

wy/profile - VPTQ, A Flexible and Extreme low-bit quantization algorithm

YangWang92 starred UDC-GAC/venom
YangWang92 starred MartaAndronic/NeuraLUT
YangWang92 starred usyd-fsalab/fp6_llm
YangWang92 starred richards199999/Thinking-Claude
YangWang92 starred NVIDIA/cuda-python
YangWang92 closed an issue on microsoft/VPTQ
When I use the parameter npercent=1 to quantize the model, I have the following problem:
/VPTQ/vptq/vptq.py", line 443, in get_error weight = weight.reshape(qweight.shape) RuntimeError: shape '[3584, 3548]' is invalid for input of size 12845056
YangWang92 created a comment on an issue on microsoft/VPTQ
Could you please list all the parameters? I suspect you might not have set vec_len for outlier. Here is an example, which allocates an outlier codebook (vector length = 4, size = 256) for **1%** o...

View on GitHub

YangWang92 starred youngsheen/SimVQ
YangWang92 starred chuanyangjin/fast-DiT
YangWang92 starred lcy-seso/pypoly
YangWang92 starred Azure/msccl-tools
YangWang92 starred NVIDIA/TensorRT-Model-Optimizer
YangWang92 starred arpitp/ssd-controller
YangWang92 starred usstq/aboutSHW
YangWang92 starred Vchitect/FasterCache
YangWang92 starred UbiquitousLearning/mllm
YangWang92 created a comment on an issue on microsoft/VPTQ
https://github.com/microsoft/VPTQ/issues/76 deduplicate

View on GitHub

YangWang92 closed an issue on microsoft/VPTQ
VLM Support
VLM Support
YangWang92 pushed 1 commit to main microsoft/VPTQ

View on GitHub

YangWang92 closed a pull request on microsoft/VPTQ
Update README.md
YangWang92 opened a pull request on microsoft/VPTQ
Update README.md
YangWang92 pushed 1 commit to patch-1 YangWang92/VPTQ

View on GitHub

YangWang92 starred google-deepmind/alphafold3
YangWang92 starred QwenLM/Qwen2.5-Coder
YangWang92 starred leptonai/leptonai
YangWang92 starred google-ai-edge/ai-edge-quantizer
Load more