Add env var to control custom Triton kernel cache and set CSGMV as default backend. (#12176)

2025-10-27 17:49:32 -07:00
parent 88596739a4
commit ce832d7034
13 changed files with 24 additions and 25 deletions
@@ -53,7 +53,7 @@ if __name__ == "__main__":
    parser.add_argument(
        "--lora-backend",
        type=str,
-        default="triton",
+        default="csgmv",
    )
    parser.add_argument(
        "--tp-size",