← All models View on HuggingFace →
M
MoonshotAI: Kimi K2.5
moonshotai/kimi-k2.5
VisionTool useJSONReasoningStreaming
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
Pricing
Input
$0.4 / 1M
Output
$1.90 / 1M
Specs
Context
262,144 tokens
Input
text, image
Output
text
Released: 2026-01
Supported parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Open weights · HuggingFace
1,371,653 downloads/mo
2,805 likes
other image-text-to-text