Clean up server_args.py to have a dedicated function for model specific adjustments (#8983)

2025-08-08 19:56:50 -07:00
parent 23f2afb2ce
commit 706bd69cc5
24 changed files with 201 additions and 340 deletions
--- a/test/srt/test_ascend_tp1_bf16.py
+++ b/test/srt/test_ascend_tp1_bf16.py
@@ -15,7 +15,7 @@ from sglang.test.test_utils import (

 TEST_MODEL_MATRIX = {
    "Qwen/Qwen2.5-7B-Instruct": {
-        "accuracy": 0.85,
+        "accuracy": 0.84,
        "latency": 150,
        "output_throughput": 30,
    },