Kimi K2 0905
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, from
- $0.57 / $2.30 per 1M
- Weekly tokens / requests
- 214.9K / 72
- Context
- 262K tokens
- Released
- Sep 5, 2025
Paste into Claude Code, Cursor, or Codex
Help me set up the Respan gateway: fetch and follow https:/
Providers
4 providers, cheapest first, priced per 1M tokens after gateway discounts.
| Provider | Input | Output |
|---|---|---|
IO.net ionet/moonshotai/Kimi-K2-Instruct-0905 | $0.57/M | $2.30/M |
Featherless featherless_ai/moonshotai/Kimi-K2-Instruct-0905 | $0.60/M | $2.50/M |
Novita novita/moonshotai/kimi-k2-0905 | $0.60/M | $2.50/M |
OpenRouter openrouter/moonshotai/kimi-k2-0905 | $0.60/M | $2.50/M |
Is a provider down? OpenRouter status
Pricing
Averaged across 4 providers, in USD per 1M tokens after gateway discounts.
- Average input
- $0.59 per 1M
- Average output
- $2.45 per 1M
- Average cache read
- $0.34 per 1M
Cost by volume
Total cost from 0 to 1T tokens on the cheapest and most expensive provider.
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 better15tokens/sbest, on Novita
- Novita15 tok/s
- OpenRouter13 tok/s
- IO.net13 tok/s
Response latency
Lower is better1.1sbest, on Novita
- Novita1.1 s
- IO.net1.6 s
- OpenRouter1.8 s
Benchmarks
6 published scores from 2 sources, each linked to where it was published.
Published scores
| Benchmark | Score | Source | Published |
|---|---|---|---|
| SWE-Bench verifiedhuggingface.co | 69.2% | huggingface.co | Sep 5, 2025 |
| SWE-Devhuggingface.co | 66.6% | huggingface.co | Sep 5, 2025 |
| SWE-Bench Multilingualhuggingface.co | 55.9% | huggingface.co | Sep 5, 2025 |
| Terminal-Bench (Terminus-2)huggingface.co | 44.5% | huggingface.co | Sep 5, 2025 |
| WeirdMLHåvard Tveit Ihle | 36.7% | Håvard Tveit Ihle | Oct 5, 2026 |
| Multi-SWE-Benchhuggingface.co | 33.5% | huggingface.co | Sep 5, 2025 |
Apps
Use Kimi K2 0905 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 0905 FAQ
Short answers, built from the data on this page.
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.
