[Feature][OP] Append Attn Support CUDA-PDL (#5072)

2025-12-24 13:28:13 +08:00 · 2025-11-17 20:47:33 +08:00
parent c2c1942db9
commit d58c1db8a0
12 changed files with 2828 additions and 2068 deletions
--- a/fastdeploy/engine/engine.py
+++ b/fastdeploy/engine/engine.py
@@ -448,6 +448,7 @@ class LLMEngine:
            "NCCL_ALGO": "Ring",
            "FLAGS_max_partition_size": int(os.getenv("FLAGS_max_partition_size", 1024)),
            "OMP_NUM_THREADS": int(os.getenv("OMP_NUM_THREADS", 3)),
+            "FD_ENABLE_PDL": envs.FD_ENABLE_PDL,
        }
        # environment variables needed by Dy2St
        variables.update(