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
Map CwmForCausalLM to llama and LlamaForCausalLM (#25611)
Signed-off-by: Jacob Kahn <jacobkahn1@gmail.com> Co-authored-by: Roger Wang <hey@rogerw.io>
Jacob Kahn committed
Sep 25, 2025 at 09:37 UTC
bc092ea873422569d2300d48a40e6a3e3ac99737
2 files changed
+4
tests/models/registry.py
+3
@@ -196,6 +196,9 @@ _TEXT_GENERATION_EXAMPLE_MODELS = {
196
trust_remote_code=True),
197
"Cohere2ForCausalLM": _HfExamplesInfo("CohereForAI/c4ai-command-r7b-12-2024", # noqa: E501
198
trust_remote_code=True),
199
+ "CwmForCausalLM": _HfExamplesInfo("facebook/cwm", # noqa: E501
200
+ trust_remote_code=True,
201
+ is_available_online=False),
202
"DbrxForCausalLM": _HfExamplesInfo("databricks/dbrx-instruct"),
203
"DeciLMForCausalLM": _HfExamplesInfo("nvidia/Llama-3_3-Nemotron-Super-49B-v1", # noqa: E501
204
trust_remote_code=True),
vllm/model_executor/models/registry.py
+1
@@ -64,6 +64,7 @@ _TEXT_GENERATION_MODELS = {
64
"ChatGLMForConditionalGeneration": ("chatglm", "ChatGLMForCausalLM"),
65
"CohereForCausalLM": ("commandr", "CohereForCausalLM"),
66
"Cohere2ForCausalLM": ("commandr", "CohereForCausalLM"),
67
+ "CwmForCausalLM": ("llama", "LlamaForCausalLM"),
68
"DbrxForCausalLM": ("dbrx", "DbrxForCausalLM"),
69
"DeciLMForCausalLM": ("nemotron_nas", "DeciLMForCausalLM"),
70
"DeepseekForCausalLM": ("deepseek", "DeepseekForCausalLM"),