[MISC] Keep chunked prefill enabled by default with long context when prefix caching is enabled (#8342)

2024-09-10 15:28:28 -07:00 · 2024-09-10 15:28:28 -07:00 · b1f3e18958
commit b1f3e18958
parent 04e7c4e771
1 changed files with 0 additions and 1 deletions
--- a/vllm/engine/arg_utils.py
+++ b/vllm/engine/arg_utils.py
@ -878,7 +878,6 @@ class EngineArgs:
                if (is_gpu and not use_sliding_window and not use_spec_decode
                        and not self.enable_lora
                        and not self.enable_prompt_adapter
-                        and not self.enable_prefix_caching
                        and not has_seqlen_agnostic_layers):
                    self.enable_chunked_prefill = True
                    logger.warning(