Update README.md

README.md

@@ -9,6 +9,8 @@ This is 2-bit quantization of [mistralai/Mixtral-8x7B-Instruct-v0.1](https://hug
 ## Model loading
 Please follow the instruction of [QuIP-for-all](https://github.com/chu-tianxiang/QuIP-for-all) for usage.
 
+As an alternative, you can use [vLLM branch](https://github.com/chu-tianxiang/vllm-gptq/tree/quip_gemv) for faster inference. QuIP has to launch like 5 kernels for each linear layer, so it's very helpful for vLLM to use cuda-graph to reduce launching overhead. BTW, If you have problem installing fast-hadamard-transform from pip, you can also install it from [source](https://github.com/Dao-AILab/fast-hadamard-transform)
+
 ## Perplexity
 Measured at Wikitext with 4096 context length
 | fp16 | 2-bit |
@@ -16,4 +18,5 @@ Measured at Wikitext with 4096 context length
 | 3.8825  | 5.2799  |
 
 ## Speed
-About 16 tokens/s at batchsize = 1 using vLLM at single A100.
+Measured with `examples/benchmark_latency.py` script at vLLM repo.
+At batch size = 1, it generates at 16.3 tokens/s with single 3090.