← All models View on HuggingFace →
Qwen: Qwen3 VL 235B A22B Instruct
qwen/qwen3-vl-235b-a22b-instruct
VisionTool useJSONStreaming
Qwen3-VL-235B-A22B Instruct is an open-weight multimodal model that unifies strong text generation with visual understanding across images and video. The Instruct model targets general vision-language use (VQA, document parsing, chart/table...
Pricing
Input
$0.2 / 1M
Output
$0.88 / 1M
Specs
Context
262,144 tokens
Input
text, image
Output
text
Knowledge cutoff: 2025-03-31
Released: 2025-09
Supported parameters
frequency_penaltylogit_biasmax_tokensmin_ppresence_penaltyrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_p
Open weights · HuggingFace
1,761,800 downloads/mo
390 likes
apache-2.0 image-text-to-text