Kimi K2 Thinking

Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in Kimi K2, it activates 32 billion parameters per forward pass and supports 256 k-token context windows. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift. It sets new open-source benchmarks on HLE, BrowseComp, SWE-Multilingual, and LiveCodeBench, while maintaining stable multi-agent behavior through 200–300 tool calls. Built on a large-scale MoE architecture with MuonClip optimization, it combines strong reasoning depth with high inference efficiency for demanding agentic and analytical tasks.

ChatMoonshot262k tokens$0.654 / $2.7251 · 1M

moonshotai/kimi-k2-thinking

Context
262k tokens
Completion cap
100,352
Tools
Yes
JSON
Yes
Released
2025-11-06

Call it from Geek Hub

Same OpenAI SDK. Change the base URL and the model id.

curl https://api.geekhub.mx/v1/chat/completions \
  -H "Authorization: Bearer $GEEKHUB_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "moonshotai/kimi-k2-thinking",
    "messages": [{"role": "user", "content": "Hola"}]
  }'
Get an API key

Frequently asked questions

What is Kimi K2 Thinking?
Kimi K2 Thinking is Moonshot AI’s most advanced open reasoning model to date, extending the K2 series into agentic, long-horizon reasoning. Built on the trillion-parameter Mixture-of-Experts (MoE) architecture introduced in Kimi K2, it activates 32 billion parameters per forward pass and supports 256 k-token context windows. The model is optimized for persistent step-by-step thought, dynamic tool invocation, and complex reasoning workflows that span hundreds of turns. It interleaves step-by-step reasoning with tool use, enabling autonomous research, coding, and writing that can persist for hundreds of sequential actions without drift. It sets new open-source benchmarks on HLE, BrowseComp, SWE-Multilingual, and LiveCodeBench, while maintaining stable multi-agent behavior through 200–300 tool calls. Built on a large-scale MoE architecture with MuonClip optimization, it combines strong reasoning depth with high inference efficiency for demanding agentic and analytical tasks. Kimi K2 Thinking runs on the Geek Hub API (OpenAI-compatible). Model id: moonshotai/kimi-k2-thinking.
Is Kimi K2 Thinking free?
No. Input is $0.654 per 1M tokens and output is $2.7251 per 1M tokens on Geek Hub (markup included).
What is the context length of Kimi K2 Thinking?
Kimi K2 Thinking has a 262k tokens context window. It supports up to 100,352 completion tokens.
Does Kimi K2 Thinking support tool calling and structured outputs?
Kimi K2 Thinking accepts tools and tool_choice for function calling. It also supports structured outputs via a JSON schema in response_format.
When was Kimi K2 Thinking released?
Kimi K2 Thinking was released on 2025-11-06.

More models from Moonshot