A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 5, 21:00 UTC
Add short flag `-sc` for `--speculative-config` argument (#38380)
Co-authored-by: Claude <noreply@anthropic.com>
Michael Goin committed
Mar 27, 2026 at 20:04 UTC
7ba425e916fe638e0dfdd64044cde17796372c95
1 file changed
+1
-1
vllm/engine/arg_utils.py
+1
-1
@@ -1301,7 +1301,7 @@ class EngineArgs:
1301
# delay the Pydantic validation that comes with SpeculativeConfig.
1302
vllm_kwargs["speculative_config"]["type"] = optional_type(json.loads)
1303
vllm_group.add_argument(
1304
- "--speculative-config", **vllm_kwargs["speculative_config"]
1304
+ "--speculative-config", "-sc", **vllm_kwargs["speculative_config"]
1305
)
1306
vllm_group.add_argument(
1307
"--kv-transfer-config", **vllm_kwargs["kv_transfer_config"]