ayo/ipex-llm
1
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions 1

Add extra warmup for THUDM/glm-4-9b-chat in igpu-performance test (#11417)

Browse source
This commit is contained in:
Yuwen Hu 2024-06-24 18:08:05 +08:00 committed by GitHub
parent ecb9efde65
commit 75f836f288
No known key found for this signature in database
GPG key ID: B5690EEEBB952194
1 changed files with 1 additions and 1 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

2
.github/workflows/llm_performance_tests.yml vendored
View file

@ -562,7 +562,7 @@ jobs:
shell: bash shell: bash
run: | run: |
sed -i '/^\s*result = run_transformer_int4_gpu_win(repo_id, local_model_hub, in_out_pairs, warm_up, num_trials, num_beams, low_bit, cpu_embedding, batch_size, streaming)/ i\ sed -i '/^\s*result = run_transformer_int4_gpu_win(repo_id, local_model_hub, in_out_pairs, warm_up, num_trials, num_beams, low_bit, cpu_embedding, batch_size, streaming)/ i\
if repo_id in ["THUDM/chatglm3-6b"]:\ if repo_id in ["THUDM/chatglm3-6b", "THUDM/glm-4-9b-chat"]:\
run_transformer_int4_gpu_win(repo_id, local_model_hub, in_out_pairs, warm_up, num_trials, num_beams, low_bit, cpu_embedding, batch_size, streaming) run_transformer_int4_gpu_win(repo_id, local_model_hub, in_out_pairs, warm_up, num_trials, num_beams, low_bit, cpu_embedding, batch_size, streaming)
' python/llm/dev/benchmark/all-in-one/run.py ' python/llm/dev/benchmark/all-in-one/run.py