Kimi K2.5Open Weight
Released
Kimi K2.5 is an AI model released by Moonshot AI on Tuesday, Jan 27 2026, 82 days after Kimi K2 Thinking. It is an open-weight model — the trained weights are available to download and run. It is a 1T parameter model with a 256k token context window. Benchmark results (shown below) cover BullshitBench v2, SWE-Bench Verified, Next.js Evals, LiveCodeBench, BrowseComp, Humanity's Last Exam, and 1 more.
Get Moonshot AI releases and AI news.
Every other Monday
API pricing
- InputWhat you pay for everything you send the model — your question, plus any documents or earlier conversation you include with it.
- $0.60
- Cached inputA reduced rate for text you send over and over. If every request starts with the same instructions or the same document, the provider keeps a copy ready and charges less to read it again.
- $0.10
- OutputWhat you pay for the text the model writes back. It is normally the dearer half: producing an answer costs more than reading one.
- $3.00
- Reasoning or thinking is supported.
- The provider says this model is unavailable to newly registered users and is scheduled to sunset on 2026-08-31.
Available from
| SiliconFlow | — | $0.45 | $2.25 | 256K | int4 |
|---|---|---|---|---|---|
| AtlasCloud | — | $0.49 | $2.50 | 256K | int4 |
| Novita | — | $0.57 | $2.85 | 256K | — |
| Amazon Bedrock | us-east-2 | $0.60 | $3.00 | 256K | — |
| Venice | — | $0.532 | $3.325 | 256K | — |
Benchmarks
Coding
SWE-Bench VerifiedCoding — Real coding tasks pulled from open-source projects — the AI has to find and fix actual bugs. A human-checked version of the original SWE-Bench. Higher is better.
76.8%
#24 of 57Best: Claude Opus 5 · 96%
Next.js EvalsNext.js coding — Vercel's open eval of how well AI coding agents build and migrate real Next.js apps — measured as the share of tasks the agent completes successfully. Higher is better.
16%
#32 of 32Best: Claude Fable 5.1 · 97%
LiveCodeBenchCompetitive coding — Coding problems published so recently the AI can't have seen them in training — a contamination-free test of raw programming skill. Higher is better.via BenchLM
85%
#8 of 12Best: DeepSeek-V4-Pro · 93.5%
Agentic & tool use
BrowseCompWeb browsing — Can the AI browse the web and track down hard-to-find answers? Higher is better.via BenchLM
60.6%
#26 of 28Best: GPT-5.6 Sol · 92.2%
Reasoning & science
Humanity's Last ExamMultidisciplinary reasoning — Humanity's Last Exam — extremely hard expert questions across many subjects. “With tools” means the AI is allowed to search the web or run code while answering. Higher is better.via BenchLM
30.1%with tools
#40 of 45Best: Claude Opus 5.5 · 67.7%
GPQA DiamondScience — Graduate-level science questions in biology, physics, and chemistry — hard enough that subject-matter PhDs score around 65%. Higher is better.
87.6%
#24 of 59Best: GPT-6 Astra · 96%
Robustness
BullshitBench v2Nonsense detection — Given a confidently-worded but nonsensical prompt, does the AI spot that it makes no sense and push back — instead of playing along and inventing an answer? The score is how often it clearly called out the nonsense. Higher is better.
52%
#31 of 84Best: Claude Opus 4.8 · 95%
Source: BenchLM, retrieved 31 August 2026. Other sources are identified on the linked benchmark pages.
Compare Kimi K2.5 with
Frequently asked questions
Kimi K2.5 was released by Moonshot AI on Tuesday, Jan 27 2026.