kimi-k2-thinking
Kimi K2 Thinking is an advanced open-source thinking model by Moonshot AI. It can execute up to 200 – 300 sequential tool calls without human interference, reasoning coherently across hundreds of steps to solve complex problems. Built as a thinking agent, it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and other benchmarks, with major gains in reasoning, agentic search, coding, writing, and general capabilities.
| Provider | Source | Input Price ($/1M) | Output Price ($/1M) | Description | Free |
|---|---|---|---|---|---|
| vercel | vercel | Input: $0.47 | Output: $2.00 | Kimi K2 Thinking is an advanced open-source thinking model by Moonshot AI. It can execute up to 200 – 300 sequential tool calls without human interference, reasoning coherently across hundreds of steps to solve complex problems. Built as a thinking agent, it reasons step by step while using tools, achieving state-of-the-art performance on Humanity's Last Exam (HLE), BrowseComp, and other benchmarks, with major gains in reasoning, agentic search, coding, writing, and general capabilities. | |
| together | together | Input: $1.20 | Output: $4.00 | - | |
| poe | poe | Input: $6,700.00 | Output: - | Built as a thinking agent, it performs step-by-step reasoning while utilizing tools, achieving state-of-the-art performance on benchmarks such as Humanity's Last Exam (HLE), BrowseComp, and others. The model demonstrates substantial advancements in reasoning, agentic search, coding, writing, and general problem-solving capabilities. Kimi K2 Thinking is capable of executing 200–300 sequential tool calls autonomously, maintaining coherent reasoning across hundreds of steps to solve complex tasks. File Support: Text, Markdown and PDF files Context window: 256k tokens | |
| moonshotaicn | models-dev | Input: $0.60 | Output: $2.50 | Provider: Moonshot AI (China), Context: 262144, Output Limit: 262144 | |
| moonshotai | models-dev | Input: $0.60 | Output: $2.50 | Provider: Moonshot AI, Context: 262144, Output Limit: 262144 | |
| nvidia | models-dev | Input: $0.00 | Output: $0.00 | Provider: Nvidia, Context: 262144, Output Limit: 262144 | |
| venice | models-dev | Input: $0.75 | Output: $3.20 | Provider: Venice AI, Context: 262144, Output Limit: 65536 | |
| siliconflowcn | models-dev | Input: $0.55 | Output: $2.50 | Provider: SiliconFlow (China), Context: 262000, Output Limit: 262000 | |
| kimiforcoding | models-dev | Input: $0.00 | Output: $0.00 | Provider: Kimi For Coding, Context: 262144, Output Limit: 32768 | |
| cortecs | models-dev | Input: $0.66 | Output: $2.73 | Provider: Cortecs, Context: 262000, Output Limit: 262000 | |
| togetherai | models-dev | Input: $1.20 | Output: $4.00 | Provider: Together AI, Context: 262144, Output Limit: 32768 | |
| azure | models-dev | Input: $0.60 | Output: $2.50 | Provider: Azure, Context: 262144, Output Limit: 262144 | |
| baseten | models-dev | Input: $0.60 | Output: $2.50 | Provider: Baseten, Context: 262144, Output Limit: 262144 | |
| siliconflow | models-dev | Input: $0.55 | Output: $2.50 | Provider: SiliconFlow, Context: 262000, Output Limit: 262000 | |
| helicone | models-dev | Input: $0.48 | Output: $2.00 | Provider: Helicone, Context: 256000, Output Limit: 262144 | |
| opencode | models-dev | Input: $0.40 | Output: $2.50 | Provider: OpenCode Zen, Context: 262144, Output Limit: 262144 | |
| zenmux | models-dev | Input: $0.60 | Output: $2.50 | Provider: ZenMux, Context: 262144, Output Limit: 64000 | |
| iflowcn | models-dev | Input: $0.00 | Output: $0.00 | Provider: iFlow, Context: 128000, Output Limit: 64000 | |
| synthetic | models-dev | Input: $0.55 | Output: $2.19 | Provider: Synthetic, Context: 262144, Output Limit: 262144 | |
| deepinfra | models-dev | Input: $0.47 | Output: $2.00 | Provider: Deep Infra, Context: 131072, Output Limit: 32768 | |
| nanogpt | models-dev | Input: $1.00 | Output: $2.00 | Provider: NanoGPT, Context: 32768, Output Limit: 8192 | |
| fireworksai | models-dev | Input: $0.60 | Output: $2.50 | Provider: Fireworks AI, Context: 256000, Output Limit: 256000 | |
| ionet | models-dev | Input: $0.55 | Output: $2.25 | Provider: IO.NET, Context: 32768, Output Limit: 4096 | |
| azurecognitiveservices | models-dev | Input: $0.60 | Output: $2.50 | Provider: Azure Cognitive Services, Context: 262144, Output Limit: 262144 | |
| moonshot | litellm | Input: $0.60 | Output: $2.50 | Source: moonshot, Context: 262144 | |
| openrouter | openrouter | Input: $0.32 | Output: $0.48 | 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. Context: 262144 |