Jonas Li comments

Repositories
Issues
Comments

Results 4 comments of


                                            Jonas Li

bert model split into many layers after int8 quantization

@DamonsJ Can you help to upload the ONNX files before and after the quantization so that we can further investigate the multi-layer splitting and the performance diff in the original...

[https://nvbugs/5670793][fix] Solve trtllm-serve launch_disaggregated issue

/bot run --disable-fail-fast

[https://nvbugs/5670793][fix] Solve trtllm-serve launch_disaggregated issue

/bot run

[https://nvbugs/5667774][fix] Refine Piecewise Cuda Graph Condition for DP

/bot run