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 12, 18:30 UTC
Fix mistral config (#29172)
Signed-off-by: Julien Denize <julien.denize@mistral.ai> Signed-off-by: Julien Denize <40604584+juliendenize@users.noreply.github.com> Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com> Co-authored-by: Patrick von Platen <patrick.v.platen@gmail.com>
Julien Denize committed
Nov 21, 2025 at 15:01 UTC
434f3d3eb869606af221f0307e16548c1f99da20
1 file changed
+4
vllm/transformers_utils/configs/mistral.py
+4
@@ -90,6 +90,10 @@ def _remap_mistral_yarn_args(config: dict) -> dict:
90
"rope_type": "yarn",
91
"mscale_all_dim": 1,
92
}
93
+
94
+ if rope_theta := config.pop("rope_theta", None):
95
+ config["rope_parameters"]["rope_theta"] = rope_theta
96
+
97
for old_name, new_name in yarn_config_map.items():
98
if old_name in yarn_config:
99
config["rope_parameters"][new_name] = yarn_config.pop(old_name)