wang fei
shanghai
Repositories
Issues
Comments
Results
3
comments of
wang fei
add Sequence Parallelism
+1
[Bug] DeepSeek R1 serve crash occasionally on 2*H100
+1
[Performance]: When deployed DeepSeek-V3 on 8*H20(96GB), maximum model length only reaches 6500 using vllm, but with sglang can achieve 163840.
+1，so on H200
