All models

Kimi-VL-A3B-Thinking-2506

byMoonshotMoonshot· 21 Jun 2025
General purposeMultimodal

An update to Kimi's visual reasoning MoE model. This version uses fewer thinking tokens, while having a superior performance over its predecessor.

Specs
Params16B
LicenseMIT
Tags
modelsmodels/llmsmodels/llms/multi-modalmodels/llms/reasoningartifacts/11
Adoption · Interconnects
RAM score
Hugging Face Downloads
4.1K
last 30d
941.6K
all time
HF Likes
366

Relative Adoption Metric contextualizes downloads against the model's size bucket.

VAIL
VAIL Fingerprint
02bd:03dc:04c8:0769:0a17:0c60:13a3:5b73

Explore other models with behavioral similarity to Kimi-VL-A3B-Thinking-2506.

Related Models