Olga Miroshnichenko
Results
1
issues of
Olga Miroshnichenko
Please consider changing framework for DeepSeek R1 to vLLM, it shows better performance over SGLang. Here is also [documentation for running DeepSeek with vLLM](https://rocm.docs.amd.com/en/docs-7.0-docker/benchmark-docker/inference-vllm-deepseek-r1-fp8.html).