biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[UX] Add --moe-backend arg for explicit kernel selection (#33807)

Browse Source 
Signed-off-by: mgoin <mgoin64@gmail.com>
Co-authored-by: Robert Shaw <114415538+robertgshaw2-redhat@users.noreply.github.com>
This commit is contained in:
Michael Goin
2026-02-25 20:44:44 -05:00
committed by GitHub
parent 1976356ee6
commit de527e1cec
37 changed files with 260 additions and 140 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
tests/evals/gsm8k/configs/moe-refactor/Llama-4-Scout-Fp8-ModelOpt-triton.yaml
View File

@@ -2,6 +2,4 @@ model_name: "nvidia/Llama-4-Scout-17B-16E-Instruct-FP8"
accuracy_threshold: 0.92
num_questions: 1319
num_fewshot: 5
server_args: "--enforce-eager --max-model-len 8192 --tensor-parallel-size 2"
env:
VLLM_USE_FLASHINFER_MOE_FP8: "0"
server_args: "--enforce-eager --max-model-len 8192 --tensor-parallel-size 2 --moe-backend=triton"