MMoonshot AI · Released Jan 27, 2026
Kimi K2.5
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
- Top 20%Fiction.LiveBench
- Released
- Jan 27 2026
- Context window
- 262K 236K out
- Input / 1M tokens
- $0.50 from $0.49
- Output / 1M tokens
- $2.50
Where to run it
5 providers compared
Cheapest: AtlasCloud at $0.49 / $2.50 per 1M input / output tokens — 19% less than the priciest option.
| Provider | Input | Output | Cache read | Context | Max output | Uptime | vs. best |
|---|---|---|---|---|---|---|---|
| AtlasCloudINT4 | $0.49 | $2.50 | $0.20 | 262K | 236K | 99%24h | Cheapest |
| SiliconFlowINT4 | $0.50 | $2.50 | $0.15 | 262K | 236K | 100%24h | +1% |
| Novita | $0.57 | $2.85 | $0.095 | 262K | 236K | 99%24h | +15% |
| Amazon Bedrockus-east-2 | $0.60 | $3.00 | — | 262K | 131K | 100%24h | +21% |
| Venice | $0.53 | $3.33 | $0.21 | 256K | 66K | 99.9%24h | +24% |
Value
Price vs. performance
Kimi K2.5 Other Moonshot AI Other labs Value frontier · SWE-bench Verified vs. input price per 1M tokens (log scale) · Drag to pan, pinch or Ctrl + scroll to zoom, double-click to zoom in.
Benchmarks
13 benchmark scores
| Benchmark | Score | Rank | Leader |
|---|---|---|---|
| SWE-bench VerifiedSoftware engineering · Real GitHub issues fixed in Python repos | 73.8% | #16of 26 | Claude Opus 4.783.5% |
| Terminal-BenchSoftware engineering · Tasks completed with the programs in a real terminal (best agent) | 43.2% | #21of 40 | GPT-5.584.7% |
| SciCodeSoftware engineering · Research-coding problems from the natural sciences | 49.0% | #52of 93 | Claude Opus 5.566.9% |
| OSWorldAgents & computer use · Desktop and web tasks with keyboard and mouse | 63.3% | #3of 4 | Claude Sonnet 4.672.1% |
| Vending-Bench 2Agents & computer use · Run a simulated vending business for a year (final balance) | $1,198 | #46of 61 | GPT-6 Astra$15,515 |
| Fiction.LiveBenchWriting & content · Understanding long creative writing (16k tokens) | 86.1% | #3of 15 | GPT-597.2% |
| SimpleQA VerifiedWriting & content · Factual accuracy on 1,000 short questions | 34.3% | #48of 66 | Gemini 4 Argon76.0% |
| WebDev ArenaDesign & visual · People vote on which model builds the better website (Elo) | 1436 | #57of 88 | Claude Opus 5.51820 |
| Humanity's Last ExamReasoning & knowledge · Expert questions across 100+ subjects | 24.4% | #15of 23 | GPT-6 Astra54.8% |
| ARC-AGI-2Reasoning & knowledge · Few-shot abstract reasoning on grid puzzles | 11.8% | #49of 63 | GPT-6 Astra95.0% |
| ARC-AGI-1Reasoning & knowledge · Pattern induction from input–output examples | 65.3% | #48of 63 | Gemini 4 Argon98.5% |
| SimpleBenchReasoning & knowledge · Common-sense trick questions about space, time and people | 46.8% | #34of 43 | Claude Fable 581.9% |
| CritPtReasoning & knowledge · Open-ended, research-level physics problems | 3.1% | #63of 96 | GPT-5.6 Sol32.3% |
Specs
Details
- Model ID
moonshotai/kimi-k2.5- Developer
- Moonshot AI
- Release date
- Jan 27, 2026 ·
- Context window
- 262,144 tokens
- Max output
- 235,929 tokens
- API price
- $0.50 / $2.50 per 1M input / output tokens
- Providers
- 5
- Input types
- text, image
- Reasoning
- Yes
- Tool calling
- Yes
- Open weights
- Yes
- Free tier
- No
Release date, prices and providers from OpenRouter's catalog; buying directly from a provider can cost a little differently. Uptime is the share of OpenRouter requests each provider served over the last day. Benchmark scores are the best result per benchmark from Epoch AI, licensed CC BY 4.0. Rank counts every model tracked here with a score.