Upload Meta-Llama-3-8B-Instruct, seqlen = 512, python, w_ compile.png
#392
by
kwen2501
- opened
Tensor parallel performance benchmark for PR https://github.com/huggingface/transformers/pull/34184.
To be used in transformers/docs/source/en/perf_infer_gpu_multi.md
ArthurZ
changed pull request status to
merged
Thanks