MoonshotAI: Kimi K3 (batch)
by Moonshot AItext+image+video->text1 endpoint
Overview
Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.
Capabilities
Text generation
Image understanding
Video understanding
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Modalities
Input
TextImageVideo
Output
Text
Technical Specifications
Context Window
1.0M tokens
Max Output
943.7K tokens
Tokenizer
Other
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biasmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatstopstructured_outputstemperaturetool_choicetoolstop_ktop_p
Pricing
Per 1M tokens
Input$3.00
Output$15.00
Cache read$0.30
Added
July 16, 2026
Last synced 9/16/2026