[Model] Support Qwen3-VL Model Series (#24727)

Signed-off-by: Roger Wang <hey@rogerw.io>
Signed-off-by: Isotr0py <mozf@mail2.sysu.edu.cn>
Co-authored-by: Huang Jie <92386084+JJJYmmm@users.noreply.github.com>
Co-authored-by: 松灵 <26085463+wulipc@users.noreply.github.com>
Co-authored-by: Isotr0py <mozf@mail2.sysu.edu.cn>
This commit is contained in:
Roger Wang
2025-09-16 22:01:04 -07:00
committed by GitHub
parent 5801e49776
commit 0f7acdd73c
13 changed files with 2084 additions and 17 deletions

View File

@@ -259,11 +259,13 @@ _MULTIMODAL_MODELS = {
"Qwen2AudioForConditionalGeneration": ("qwen2_audio", "Qwen2AudioForConditionalGeneration"), # noqa: E501
"Qwen2_5OmniModel": ("qwen2_5_omni_thinker", "Qwen2_5OmniThinkerForConditionalGeneration"), # noqa: E501
"Qwen2_5OmniForConditionalGeneration": ("qwen2_5_omni_thinker", "Qwen2_5OmniThinkerForConditionalGeneration"), # noqa: E501
"UltravoxModel": ("ultravox", "UltravoxModel"),
"Qwen3VLForConditionalGeneration": ("qwen3_vl", "Qwen3VLForConditionalGeneration"), # noqa: E501
"Qwen3VLMoeForConditionalGeneration": ("qwen3_vl_moe", "Qwen3VLMoeForConditionalGeneration"), # noqa: E501
"SkyworkR1VChatModel": ("skyworkr1v", "SkyworkR1VChatModel"),
"Step3VLForConditionalGeneration": ("step3_vl", "Step3VLForConditionalGeneration"), # noqa: E501
"TarsierForConditionalGeneration": ("tarsier", "TarsierForConditionalGeneration"), # noqa: E501
"Tarsier2ForConditionalGeneration": ("qwen2_vl", "Tarsier2ForConditionalGeneration"), # noqa: E501
"UltravoxModel": ("ultravox", "UltravoxModel"),
"VoxtralForConditionalGeneration": ("voxtral", "VoxtralForConditionalGeneration"), # noqa: E501
# [Encoder-decoder]
"WhisperForConditionalGeneration": ("whisper", "WhisperForConditionalGeneration"), # noqa: E501