Moonshot AIKimi K2.7 Code
Kimi K2.7 Code is a Moonshot AI open-weights reasoning model, released 12 Jun 2026. Its best configuration ranks #221 of 785 on the BenchLeader Index at 55.8 ±4.1, in the upper half. The ± is the point: 268 other configurations score within that range, so they and this one cannot be told apart on quality alone — price and speed are what separate them. It scores highest in long context (65) and lowest in knowledge (51). At $1.71 per million tokens blended it is pricier than most ranked models. Output speed of 54 tokens per second puts it slower than most, with a first answer in 44.4 s. It has been measured at 2 reasoning-effort settings; this summary describes the best-scoring one, and the tabs above switch between them. Last measured 22 Oct 2026.
- Blended price
- $1.71/M
- $0.950 in · $4.00 out
- Output speed
- 54 tok/s
- measured by Artificial Analysis
- First answer
- 44 s
- first token 0.81 s
- Context
- 262k
- Full answer
- 54 s
- median, reasoning included
- Cost per run
- $0.541
- one full Intelligence Index run
- Released
- 12 Jun 2026
- Overall index56
- Reasoning59
- Coding52
- Agents & tools52
- Maths53
- Knowledge51
- Instruction following63
- Long context65
- Composite58
Reasoning-effort configurations
The same model behaves differently depending on how much it is allowed to think. Each row is one setting, scored only on the benchmarks that were run at that setting. “Not stated” collects results from publishers that did not say which setting they used; for a reasoning model that is usually its thinking mode, but we do not assume it. Pick a setting here or at the top of the page to see its price, speed and category scores.
| Effort | Index | Rank | Speed | First answer | Chat reply cost | Categories |
|---|---|---|---|---|---|---|
| xhigh | – | – | 54 tok/s | 44 s | $0.0016 | |
| not statedbest | 55.8 | #221 | 54 tok/s | 44 s | $0.0016 | Agents & tools 52 · Coding 52 · Composite 58 · Instruction following 63 · Knowledge 51 · Long context 65 · Maths 53 · Reasoning 59 |
How its index has moved
8 Sept 2026 to 21 Oct 2026The index is recomputed from scratch every day, so a line moves when a new benchmark result lands, when a publisher revises a score, or when the models it is normalised against change. Early movement usually means the score is still settling.
Benchmark results
See all LLM benchmarks. One column per reasoning effort. Rank is among every configuration of every model on that benchmark. Hover a score for the run it came from.
Reasoning
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| GPQA Diamond | 87.9%#65 | Epoch AI Benchmarking Hub | |
| SimpleBench | 57.9%#34 | SimpleBench | |
| LiveBench 1.0 Reasoningnot in index | 82.8%#45 | LiveBench | |
| GPQA Diamond (AA)not in index | 89.6%#71 | Artificial Analysis | |
| Humanity's Last Exam (AA)not in index | 35.0%#121 | Artificial Analysis | |
| CritPt | 10.0%#121 | Artificial Analysis | |
| Chess Puzzlesnot in index | 14.0%#111 | Epoch AI Benchmarking Hub |
Coding
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| SciCode | 47.8%#123 | SciCode | |
| WeirdML | 54.1%#61 | WeirdML | |
| FrontierCode | 30.1%#32 | Cognition | |
| Arena WebDev | 1474#67 | Arena (LMArena) | |
| LiveBench 1.0 Codingnot in index | 74.0%#54 | LiveBench | |
| SciCode (AA)not in index | 47.8%#129 | Artificial Analysis | |
| LiveCodeBench | 82.0%#56 | Vals AI | |
| SWE-bench (Vals)not in index | 78.2%#30 | Vals AI | |
| Code Migrationnot in index | 25.4%#47 | Vals AI | |
| ProgramBenchnot in index | 0.0%#32 | Vals AI | |
| Vibe Code Bench v1.1not in index | 47.2%#66 | Vals AI | |
| Surface Evolver Benchnot in index | 48.8%#19 | Epoch AI Benchmarking Hub | |
| DeepSWE v1.1not in index | 30.5%#62 | Epoch AI Benchmarking Hub | |
| ALE-Benchnot in index | 886.2#68 | Epoch AI Benchmarking Hub | |
| Terminal-Bench 4.0 (AA)not in index | 1.0%#128 | Artificial Analysis | |
| Terminal-Bench 2.1 (AA)not in index | 67.4%#78 | Artificial Analysis |
Agents & tools
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| APEX-Agents | 37.6%#45 | Mercor | |
| LiveBench Agentic Codingnot in index | 45.7%#53 | LiveBench | |
| Terminal-Bench Hard | 44.7%#36 | Artificial Analysis | |
| τ²-Bench Telecom (AA)not in index | 90.1%#65 | Artificial Analysis | |
| Terminal-Bench 2.1 (Vals) | 67.0%#36 | Vals AI | |
| GDPval-AA v2.1not in index | 26.9%#91 | Artificial Analysis | |
| τ³-Banking (AA) | 20.2%#98 | Artificial Analysis | |
| SkillsBenchnot in index | 50.0%#22 | Vals AI | |
| GBAEvalnot in index | 0.8%#18 | Epoch AI Benchmarking Hub | |
| Vending-Bench 2not in index | 5082.9#32 | Andon Labs | |
| AutomationBenchnot in index | 24.4%#127 | Artificial Analysis | |
| GDP.pdf (AA)not in index | 11.2%#120 | Artificial Analysis | |
| EnterpriseOps-Gymnot in index | 40.2%#29 | Artificial Analysis |
Maths
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| FrontierMath Tiers 1–3 | 54.0%#51 | Epoch AI Benchmarking Hub | |
| FrontierMath Tier 4 | 12.2%#62 | Epoch AI Benchmarking Hub | |
| OTIS Mock AIME | 95.6%#44 | Epoch AI Benchmarking Hub | |
| LiveBench Mathematicsnot in index | 79.6%#61 | LiveBench |
Knowledge
| Benchmark | xhigh | not stated | Source | Trend |
|---|---|---|---|---|
| SimpleQA Verifiedeffort not stated | – | 36.5%#50 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0 Data Analysisnot in indexeffort not stated | – | 62.7%#60 | LiveBench | |
| AA-Omniscienceeffort not stated | – | -10.2#193 | Artificial Analysis | |
| DrugDiscoveryBenchnot in index | 35.3%#13 | – | Scale AI SEAL | |
| AA-Omniscience: accuracynot in indexeffort not stated | – | 39.6%#127 | Artificial Analysis | |
| AA-Omniscience: non-hallucinationnot in indexeffort not stated | – | 17.6%#311 | Artificial Analysis |
Instruction following
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| LiveBench Languagenot in index | 77.9%#40 | LiveBench | |
| LiveBench Instruction Followingnot in index | 56.3%#62 | LiveBench | |
| IFBench | 63.1%#116 | Artificial Analysis |
Long context
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| AA-LCR | 79.3%#98 | Artificial Analysis | |
| MLCRnot in index | 12.2%#96 | Artificial Analysis |
Composite
| Benchmark | not stated | Source | Trend |
|---|---|---|---|
| Epoch Capabilities Indexnot in index | 150.1#57 | Epoch AI Benchmarking Hub | |
| LiveBench 1.0not in index | 68.4%#58 | LiveBench | |
| AA Intelligence Index v4.3.2 | 25.8#160 | Artificial Analysis |
Where to run it
Every provider serving this model through OpenRouter, with throughput and first-token latency measured on live traffic over the last 30 minutes and each provider’s own price. Purple marks the best in each column.
| Provider | Speed | First token | Input $/M | Output $/M | Context | Quantisation |
|---|---|---|---|---|---|---|
| Moonshot AI | 105 tok/s | 0.94 s | $1.90 | $8.00 | 262k | int4 |
| Alibaba Cloud Int. | 69 tok/s | 1.27 s | $0.950 | $4.00 | 262k | fp8 |
| Inceptron | 62 tok/s | 0.84 s | $0.671 | $3.35 | 262k | int4 |
| Cloudflare | 56 tok/s | 0.82 s | $0.950 | $4.00 | 262k | – |
| StreamLake | 51 tok/s | 1.30 s | $0.713 | $3.00 | 256k | – |
| Moonshot AI | 48 tok/s | 1.06 s | $0.950 | $4.00 | 262k | int4 |
| NovitaAI | 38 tok/s | 2.31 s | $0.912 | $3.84 | 262k | int4 |
| Venice | 33 tok/s | 3.26 s | $0.750 | $3.50 | 256k | int4 |
| SiliconFlow | 29 tok/s | 1.15 s | $0.859 | $3.80 | 262k | fp8 |
| Nebius Token Factory | 24 tok/s | 1.49 s | $0.950 | $4.00 | 262k | fp4 |
| CoreWeave | 15 tok/s | 3.76 s | $0.710 | $3.50 | 262k | int4 |
What a task costs
Estimates from list price, output speed and time to first answer for the best configuration. “With caching” assumes three-quarters of the input is served from the prompt cache at $0.190 per 1M. Reasoning tokens are not modelled.
| Workload | Tokens in / out | Cost | With caching | Time |
|---|---|---|---|---|
| Chat reply | 400 / 300 | $0.0016 | $0.0014 | 50.0 s |
| Summarize a 30-page report | 12,000 / 600 | $0.014 | $0.0070 | 55.6 s |
| Code edit | 6,000 / 1,500 | $0.012 | $0.0083 | 1.2 min |
| Agentic coding session | 60,000 / 4,000 | $0.073 | $0.039 | 2.0 min |
| Structured extraction | 2,000 / 200 | $0.0027 | $0.0016 | 48.1 s |
More from Moonshot AI
See also
Questions people ask
- What are Kimi K2.7 Code's benchmark scores?
- As of 24 Oct 2026, Kimi K2.7 Code scores 55.8 ±4.1 on the BenchLeader Index, strongest in long context (65) and weakest in knowledge (51). That rests on 20 independent benchmark results, and every score on this page links to its publisher.
- How does Kimi K2.7 Code rank?
- Kimi K2.7 Code is #221 of 785 ranked configurations on 24 Oct 2026. 268 other configurations score within its margin, so the exact position matters less than the band. The leader is Claude Opus 5.5 (max effort) at 72.6.
- How much does Kimi K2.7 Code cost per 1M tokens?
- As of 24 Oct 2026, Kimi K2.7 Code costs $0.950 per 1M input tokens and $4.00 per 1M output tokens, $1.71 blended (three input tokens per output token).
- When was Kimi K2.7 Code released?
- Moonshot AI released Kimi K2.7 Code on 12 Jun 2026. BenchLeader has results for it at 2 reasoning-effort settings.
Data as of 24 Oct 2026. Compare these configurations.
Cite as: BenchLeader, “Kimi K2.7 Code: benchmarks, pricing, speed and rank”, data as of 24 Oct 2026, https://www.benchleader.com/models/kimi-k2-7-code.