mirror of
https://git.datalinker.icu/vllm-project/vllm.git
synced 2025-12-10 10:06:03 +08:00
a22cdea371
Provide an initial support to FP8 computation. This PR is inspired by HuggingFace TGI: huggingface/text-generation-inference#1726 This feature can be enabled with --quantization fp8 or -q fp8 when launching an engine. Algorithm: We still load a model checkpoint in FP16/BF16. After the weights are loaded, Fp8LinearMethod calculates the per-tensor scaling factor of weights and quantizes the weights accordingly. The scaling factor will then be stored for future use. Meanwhile, the per-tensor scaling factor for activations is calculated in every forward pass. Initial Results: Currently tested Mistral-7B on 1xH100. With prompt length ~5 and decoding length 128: BF16: 1.47s FP8: 1.66s I'll try to use larger models and try to find more performance bottleneck. Meanwhile, you're welcome to try this code.