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 14, 05:30 UTC
Small fix for Command-A-Vision (#23268)
Signed-off-by: donglu <donglu@cohere.com>
dongluw committed
Aug 20, 2025 at 14:15 UTC
ebe56a0064f7a72a5c51d4cd6bcca165590c5bed
1 file changed
+1
-1
vllm/model_executor/models/cohere2_vision.py
+1
-1
@@ -348,7 +348,7 @@ class Cohere2VisionForConditionalGeneration(nn.Module, SupportsMultiModal,
348
vllm_config=vllm_config,
349
hf_config=config.text_config,
350
prefix=maybe_prefix(prefix, "language_model"),
351
- architectures=["Cohere2ForCausalLM"])
351
+ architectures=config.text_config.architectures)
352
353
@property
354
def dtype(self):