[refactor] Move model-capability adjustments into the resolution pipeline (#30299)

2026-07-07 21:26:55 -07:00
parent d4963f5c55
commit b14f7b4f75
29 changed files with 478 additions and 209 deletions
@@ -1355,7 +1355,9 @@ async def update_weight_version(
    # since weight_version update is a simple operation that doesn't affect model weights
    try:
        # Update the weight version in server args (the single source of truth)
-        _global_state.tokenizer_manager.server_args.weight_version = obj.new_version
+        _global_state.tokenizer_manager.server_args.override(
+            "http.update_weight_version", weight_version=obj.new_version
+        )

        return ORJSONResponse(
            {