Submit

Chinese multimodal (vision) models

Accepts image + text inputs and reasons over them. Chinese multimodal leaders: Qwen-VL, GLM-4V, InternVL. Typical uses: OCR, screenshot understanding, chart / diagram reasoning.

ModelCreatorContextOpen weightTags
Qwen VL MaxAlibaba Cloud131KYeschat · vision · ocr
Qwen VL PlusAlibaba Cloud131KYes
Qwen2.5 VL 72B InstructAlibaba Cloud32KYes
Qwen3.5-122B-A10BAlibaba Cloud262KYeschat · vision · tool_calling
Qwen3.5-27BAlibaba Cloud262KYes
Qwen3.5-35B-A3BAlibaba Cloud262KYes
Qwen3.5 397B A17BAlibaba Cloud262KYeschat · reasoning · code
Qwen3.5-9BAlibaba Cloud262KYes
Qwen3.5-FlashAlibaba Cloud1000K—
Qwen3.5 Plus 2026-02-15Alibaba Cloud1000KYes
Qwen3.5 Plus 2026-04-20Alibaba Cloud1000K—
Qwen3.6 27BAlibaba Cloud262KYes
Qwen3.6 35B A3BAlibaba Cloud262KYes
Qwen3.6 FlashAlibaba Cloud1000K—
Qwen3.6 PlusAlibaba Cloud1000KYes
Qwen3 VL 235B A22B InstructAlibaba Cloud262KYes
Qwen3 VL 235B A22B ThinkingAlibaba Cloud131KYes
Qwen3 VL 30B A3B InstructAlibaba Cloud131KYes
Qwen3 VL 30B A3B ThinkingAlibaba Cloud131KYes
Qwen3 VL 32B InstructAlibaba Cloud131KYes
Qwen3 VL 8B InstructAlibaba Cloud131KYes
Qwen3 VL 8B ThinkingAlibaba Cloud131KYes
Seed 1.6ByteDance262KYes
Seed 1.6 FlashByteDance262KYes
Seed-2.0-LiteByteDance262K—
Seed-2.0-MiniByteDance262KYes
UI-TARS 7BByteDance128KYes
MiniMax-01MiniMax1000KYes
Kimi K2.5Moonshot AI262KYeschat · reasoning · code
Kimi K2.6Moonshot AI262KYeschat · reasoning · code
MoonshotAI Kimi LatestMoonshot AI262K—
GLM 4.5VZhipu AI66KYes
GLM 4.6VZhipu AI131KYes
GLM 5V TurboZhipu AI203KYes

Other capabilities