moonshotai/Kimi-VL-A3B-Thinking-2506
Image-Text-to-Text • 16B • Updated • 8.5k • 373
Kimi-VL-A3B
Moonshot's efficient MoE VLMs, exceptional on agent, long-context, and thinking
-
-
moonshotai/Kimi-VL-A3B-Instruct
Image-Text-to-Text • 16B • Updated • 388k • 277 -
moonshotai/Kimi-VL-A3B-Thinking
Image-Text-to-Text • 16B • Updated • 171k • 450 -
moonshotai/MoonViT-SO-400M
Image Feature Extraction • 0.4B • Updated • 3.84k • 90 -
Kimi-VL Technical Report
Paper • 2504.07491 • Published • 142