[Refactor] Add -fp4-gemm-backend to replace SGLANG_FLASHINFER_FP4_GEMM_BACKEND (#16534)

Co-authored-by: Vincent Zhong <207368749+vincentzed@users.noreply.github.com>
2026-01-18 23:25:46 +08:00
parent f3a7c7dcd9
commit 4df74eb576
9 changed files with 144 additions and 18 deletions
@@ -35,7 +35,7 @@ suites = {
        TestFile("test_deepseek_v3_fp4_4gpu.py", 1500),
        TestFile("test_fp8_blockwise_gemm.py", 280),
        TestFile("test_gpt_oss_4gpu.py", 700),
-        TestFile("test_llama31_fp4.py", 90),
+        TestFile("test_nvfp4_gemm.py", 360),
    ],
    # "per-commit-8-gpu-b200": [
    #     TestFile("test_mistral_large3_basic.py", 275),  # Moved to nightly - large model