Support --enable-llama4-multimodal (#5254)

2025-04-11 04:24:14 -04:00
parent aee62d744b
commit 038bc5d521
7 changed files with 32 additions and 5 deletions
--- a/python/sglang/srt/model_executor/model_runner.py
+++ b/python/sglang/srt/model_executor/model_runner.py
@@ -281,7 +281,6 @@ class ModelRunner:
                f"Automatically reduce --mem-fraction-static to {self.mem_fraction_static:.3f} "
                f"because this is a multimodal model."
            )
-
            logger.info(
                "Automatically turn off --chunked-prefill-size for multimodal model."
            )