Update docs (#1839)

2024-10-30 02:49:08 -07:00
parent 539df95d2c
commit b548801ddb
11 changed files with 165 additions and 198 deletions
--- a/python/sglang/srt/model_executor/model_runner.py
+++ b/python/sglang/srt/model_executor/model_runner.py
@@ -124,7 +124,7 @@ class ModelRunner:
                "Automatically turn off --chunked-prefill-size and adjust --mem-fraction-static for multimodal models."
            )
            server_args.chunked_prefill_size = None
-            server_args.mem_fraction_static *= 0.95
+            self.mem_fraction_static *= 0.95
            # TODO: qwen2-vl does not support radix cache now, set disable_radix_cache=True automatically
            if self.model_config.hf_config.architectures == [
                "Qwen2VLForConditionalGeneration"