Kimi K2 Thinking
Moonshot AI · Open weights · Released Nov 6, 2025 · moonshotai/kimi-k2-thinking
Best $0.60 in / $2.50 out via Novita · bf16
262K contextToolsReasoningStructured outputPrompt cachingOpen weights
Available from 1 providerUSD / MTOK · PRICES 16 MINUTES AGO · SPEED 3 DAYS AGO
EST. MONTHLY COSTclick a row for detail · click headers to sort
| PROVIDER | INPUT | OUTPUT | CACHE | CONTEXT | QUANT | TTFT | TPS | UPTIME |
|---|---|---|---|---|---|---|---|---|
| $0.60 | $2.50 | $0.15 | 262K | bf16 | 11.7s | 40 tok/s | 100.0% |
TTFT/TPS: median (P50) over trailing 504h · chat10k profile (~10K in / 200 out) · measured via OpenRouter routing · methodology v1.1 · measured 3 days ago
SERVE THIS MODEL AND NOT LISTED? LIST YOUR INFERENCE →
Pricing variants · USD / MTok in / out
STANDARD
$0.60 / $2.50
per MTok in / out
CACHE READ $0.15
Recent changesRSS ↗
Nov 5, 2026endpoint removed · cheaperinference2 → 1
Oct 29, 2026endpoint removed · google3 → 2
Oct 28, 2026endpoint added · cheaperinference2 → 3
Oct 27, 2026endpoint removed · cheaperinference3 → 2
Oct 23, 2026endpoint added · cheaperinference2 → 3
About
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...
Specifications
Context window262,144
Max output98,304
Modalities intext
Modalities outtext
LicenseOpen weights
ReleasedNov 6, 2025
TokenizerOther
WeightsHugging Face ↗
Related
Embed badge
[](https://modelindex.ai/models/moonshotai/kimi-k2-thinking)Sources
OpenRouter API16 minutes ago
AGGREGATORHugging Face16 minutes ago
COMMUNITYModelIndex benchmarks3 days ago
MEASUREDVerified 16 minutes ago · list prices, not negotiated rates.