biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Add the support for the qwen3 next model (a hybrid attention model). (#24526)

Browse Source 
Signed-off-by: Tao He <linzhu.ht@alibaba-inc.com>
Co-authored-by: Jee Jee Li <pandaleefree@gmail.com>
This commit is contained in:
Tao He
2025-09-11 15:32:09 +08:00
committed by GitHub
parent 2048c4e379
commit e93f4cc9e3
29 changed files with 2476 additions and 61 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/transformers_utils/config.py
View File

@@ -79,7 +79,7 @@ _CONFIG_REGISTRY: dict[str, type[PretrainedConfig]] = LazyConfigDict(
ultravox="UltravoxConfig",
step3_vl="Step3VLConfig",
step3_text="Step3TextConfig",
)
qwen3_next="Qwen3NextConfig")
_CONFIG_ATTRS_MAPPING: dict[str, str] = {
"llm_config": "text_config",