Skip to main content

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.

Input / output price
$0.60 / $2.50 per 1M
Context
262K tokens
Released
Nov 6, 2025
Agent setup

Paste into Claude Code, Cursor, or Codex

Help me set up the Respan gateway: fetch and follow https://www.respan.ai/docs/agent-setup.md. Use model "kimi-k2-thinking".

Read docs

Providers

5 providers, cheapest first, OpenRouter last, priced per 1M tokens after gateway discounts.

ProviderInputOutput
Bedrockbedrock/kimi-k2-thinking
$0.60/M$2.50/M
Concentrateconcentrate/kimi-k2-thinking
$0.60/M$2.50/M
IO.netionet/kimi-k2-thinking
$0.60/M$2.50/M
Novitanovita/kimi-k2-thinking
$0.60/M$2.50/M
OpenRouteropenrouter/kimi-k2-thinking
$0.60/M$2.50/M

Is a provider down? AWS Bedrock status, OpenRouter status

Pricing

Averaged across 5 providers, in USD per 1M tokens after gateway discounts.

Average input
$0.60 per 1M
Average output
$2.50 per 1M
Average cache read
$0.20 per 1M

Cost by volume

Total cost from 0 to 1T tokens.

$0.60/MBedrock, Concentrate, IO.net, Novita, and OpenRouter

Throughput and latency

30-day averages from requests through the Respan gateway. Under each figure, every model on the gateway by its average, this one in blue.

Output speed

Higher is better

30tokens/sacross providers

Faster than 50% of the other 599 models on the gateway

Response latency

Lower is better

4.5sacross providers

Faster than 34% of the other 780 models on the gateway

Benchmarks

18 published scores from 4 sources, each linked to where it was published.

Published scores

BenchmarkScore
LiveCodeBenchV6 (no tools)huggingface.co83.1%
IMO-AnswerBench (no tools)huggingface.co78.6%
SWE-bench Verified (w/ tools)huggingface.co71.3%
SWE-bench Multilingual (w/ tools)huggingface.co61.1%
Terminal-Bench (w/ simulated tools (JSON))huggingface.co47.1%
SciCode (no tools)huggingface.co44.8%
WeirdMLHåvard Tveit Ihle42.8%
Multi-SWE-bench (w/ tools)huggingface.co41.9%

Apps

Use Kimi K2 Thinking in the tools you already run. Each guide points the app at the Respan gateway.

Setup guides for more apps are in the integration docs.

Kimi K2 Thinking FAQ

Short answers, built from the data on this page.

Kimi K2 Thinking is a model from Moonshot, released on Nov 6, 2025. 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.

On the Respan gateway, Kimi K2 Thinking costs $0.60 per 1M input tokens and $2.50 per 1M output tokens on every provider, after gateway discounts.

Kimi K2 Thinking is available through 5 providers: Bedrock, Concentrate, IO.net, Novita, and OpenRouter.

Kimi K2 Thinking accepts up to 262K tokens of context on its largest provider route.

Point any OpenAI-compatible SDK at https://gateway.respan.ai/v1 with your Respan API key and set the model to "kimi-k2-thinking". To set it up with a coding agent, copy the agent setup prompt at the top of this page.

Kimi K2 Thinking has 18 published benchmark scores, including LiveCodeBenchV6 (no tools) 83.1%, BrowseComp-ZH (w/ tools) 62.3%, and GPQA (no tools) 84.5%. Each score links to where it was published; the data is collected by Respan.

Certified where it counts.

Built to meet the security and privacy standards that enterprise and healthcare teams require.

ISO 27001

The internationally recognized standard for information security management.

SOC 2

Secure, compliant management of your data across all of our systems.

GDPR

Operated under GDPR, the world's strictest standard for data privacy.

HIPAA

HIPAA compliant, with a BAA available for healthcare teams.

Works with your entire stack

Use Respan with your favorite frameworks and tools.

AnthropicAI21 LabsxAIVercel AI SDK
LangChainOpenRouterAssemblyAICursor
LiteLLMGroqBasetenMem0
CohereCogneeLlamaIndexFireworks
DeepSeekLinkupHaystackTogether AI
PerplexityInference.netPostHogRubyLLM
SlackMastraAzure OpenAIMistral
AWS BedrockMoonshotHyperspellBAML
ModaInstructorGoogle GeminiParasail
ReductoZapierAgnoNebius
OpenAINovita AIReplicateSuperagent

Built for AI agents.
Break less.
Ship more.