Kimi K2.7 Code

Composite indexes

• Artificial Analysis Intelligence Index#57 of 151
  26independentv4.3.2, reasoningArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Claude Opus 5.5 at 58

Coding

• Terminal-Bench 4.0#52 of 96
  1%independentreasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Claude Sonnet 5.5 at 70.6%

• Terminal-Bench 2.x#40 of 105
  67.4%independentTerminal-Bench 2.1, reasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Claude Fable 5.1 at 91.4%

• SciCode#54 of 100
  47.8%independentreasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Claude Opus 5.5 at 66.9%

Agents and tool use

• τ²-Bench#20 of 89
  90.1%independenttelecom, reasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: GLM-5.2 at 99.1%

• GDPval-AA#57 of 109
  1025independentGDPval-AA v2.1, reasoningArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Muse Wonder Full at 1857

• MCP Atlas#5 of 10
  76%vendorKimi K2.7 Code model card (Hugging Face) ↗

  Leader: Kimi K3 at 84.2%

Reasoning and knowledge

• GPQA Diamond#43 of 144
  89.6%independentreasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: GPT-6 Astra at 96.1%

• Humanity’s Last Exam#51 of 153
  35%independentreasoning, no tools, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Claude Sonnet 5.5 at 64.5%

• CritPt#45 of 144
  10%independentreasoning, AA-runArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Muse Wonder Full at 33%

Vision and long context

• AA Long Context Reasoning#43 of 144
  79.3%independentreasoning, AA-LCRArtificial Analysis: Kimi K2.7 Code ↗

  Leader: Kimi K3 at 88.7%

Human preference

• LMArena WebDev#45 of 58
  1472independentarena entry "kimi-k2.7-code", 4,684 votes, checked 2026-09-25Arena (LMArena) WebDev/Code leaderboard ↗

  Leader: Claude Opus 5.5 at 1818

Other published results