RLTasksby ulam.ai
Models / Kimi

Kimi / MODEL EVIDENCE

Kimi K3

87 benchmarks · 170 observations · 87 sources · released 2026-07-16

Scores at a glance

87 benchmarks · large dot: Kimi K3; small dots: closest models where reported

AA-3D ↗981.78 Elo
572.551198.83

Kimi K3 (Max) · AA-3D · Elo ↗

GPQA Diamond ↗93.54%
87.1796.06

Kimi K3 (Max) · GPQA Diamond · AA implementation ↗

4 more reported results

93.12% · Kimi K3 (max) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗

91.92% · Kimi K3 (high) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗

84.85% · Kimi K3 (low) · GPQA diamond · Best score (across scorers) · % · Epoch AI · GPQA diamond ↗

92.93 % · Kimi K3 · GPQA Diamond · Accuracy · Vals AI · GPQA Diamond ↗

HealthBench ↗53.24%
50.765.98

Kimi K3 (Max) · HealthBench · HealthBench Professional · length-adjusted ↗

ErdősBench ↗2.740 / 4
2.113.25

Kimi K3 max · ErdősBench · mean progress / 4 · API ↗

LMCA ↗52.74%
2.7868.26

Kimi K3 (max) · LMCA · Score · % ↗

DTBench ↗91.2%
41.6598.93

Kimi K3 (max) · DTBench · Accuracy · % ↗

SAFE-Teen ↗44%
1556

Kimi K3 · Urgent safety · share of applicable ratings marked fail ↗

2 more reported results

61% · Kimi K3 · Health advice · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗

63% · Kimi K3 · Relationship boundaries · share of applicable ratings marked fail · Vals AI · SAFE-Teen ↗

Kimi Code Bench 2.0 ↗72.9%
64.276.9

Kimi K3 (max) · Kimi Code Bench 2.0 · score · Kimi K3 · official model-card comparison ↗

GDPx v1.0 ↗3.7%
1.6712.03

Kimi K3 · GDPx v1.0 · Overall · intended grading ↗

53 more reported results

3.1% · Kimi K3 · GDPx v1.0 · Eight-domain subset · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.5% · Kimi K3 · GDPx v1.0 · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.1% · Kimi K3 · GDPx v1.0 · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

11.1% · Kimi K3 · GDPx v1.0 · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.1% · Kimi K3 · GDPx v1.0 · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

9.0% · Kimi K3 · GDPx v1.0 · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.3% · Kimi K3 · GDPx v1.0 · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.7% · Kimi K3 · GDPx v1.0 · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

8.8% · Kimi K3 · GDPx v1.0 · Overall · partial credit · GDPx v1.0 — SemiAnalysis ↗

8.7% · Kimi K3 · GDPx v1.0 · Eight-domain subset · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.5% · Kimi K3 · GDPx v1.0 · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.7% · Kimi K3 · GDPx v1.0 · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

12.4% · Kimi K3 · GDPx v1.0 · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

14.5% · Kimi K3 · GDPx v1.0 · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

26.0% · Kimi K3 · GDPx v1.0 · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

16.4% · Kimi K3 · GDPx v1.0 · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

9.0% · Kimi K3 · GDPx v1.0 · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.3% · Kimi K3 · GDPx v1.0 · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.9% · Kimi K3 · GDPx v1.0 · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Aria · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Arcverex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.2% · Kimi K3 · GDPx v1.0 · Ergodex · Chip Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

11.1% · Kimi K3 · GDPx v1.0 · Arithmetic · Cybersecurity · intended grading · GDPx v1.0 — SemiAnalysis ↗

5.1% · Kimi K3 · GDPx v1.0 · LatchBio · Biology · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Living Context · Finance · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Chakra · Legal · intended grading · GDPx v1.0 — SemiAnalysis ↗

9.0% · Kimi K3 · GDPx v1.0 · Third Moment · Quant Trading · intended grading · GDPx v1.0 — SemiAnalysis ↗

7.5% · Kimi K3 · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.3% · Kimi K3 · GDPx v1.0 · Beautiful Machines · Hardware Design · intended grading · GDPx v1.0 — SemiAnalysis ↗

3.2% · Kimi K3 · GDPx v1.0 · ValenceForge · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

8.2% · Kimi K3 · GDPx v1.0 · Emulated · Materials Science · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Fleet · Alignment · intended grading · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Datacurve · Software Engineering · intended grading · GDPx v1.0 — SemiAnalysis ↗

2.0% · Kimi K3 · GDPx v1.0 · Aria · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Arcverex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.2% · Kimi K3 · GDPx v1.0 · Ergodex · Chip Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

12.4% · Kimi K3 · GDPx v1.0 · Arithmetic · Cybersecurity · partial credit · GDPx v1.0 — SemiAnalysis ↗

14.5% · Kimi K3 · GDPx v1.0 · LatchBio · Biology · partial credit · GDPx v1.0 — SemiAnalysis ↗

26.0% · Kimi K3 · GDPx v1.0 · Living Context · Finance · partial credit · GDPx v1.0 — SemiAnalysis ↗

16.4% · Kimi K3 · GDPx v1.0 · Chakra · Legal · partial credit · GDPx v1.0 — SemiAnalysis ↗

9.0% · Kimi K3 · GDPx v1.0 · Third Moment · Quant Trading · partial credit · GDPx v1.0 — SemiAnalysis ↗

7.5% · Kimi K3 · GDPx v1.0 · Extrapolate · Recursive Self-Improvement · partial credit · GDPx v1.0 — SemiAnalysis ↗

2.3% · Kimi K3 · GDPx v1.0 · Beautiful Machines · Hardware Design · partial credit · GDPx v1.0 — SemiAnalysis ↗

5.5% · Kimi K3 · GDPx v1.0 · ValenceForge · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

10.4% · Kimi K3 · GDPx v1.0 · Emulated · Materials Science · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Fleet · Alignment · partial credit · GDPx v1.0 — SemiAnalysis ↗

0.0% · Kimi K3 · GDPx v1.0 · Datacurve · Software Engineering · partial credit · GDPx v1.0 — SemiAnalysis ↗

Closest models by score

Similar positions on shared benchmark leaderboards. Smaller gaps are closer.

20.0 percentile-point gap

36 of 59 shared benchmark comparisons

21.5 percentile-point gap

36 of 59 shared benchmark comparisons

How similarity is measured

Mean absolute gap in percentile positions on shared benchmark settings; lower is closer. At least 60% overlap is required, with a minimum of three when available. Independent evaluations are used when there are at least three; otherwise published report comparisons are used. Repeated model observations within one setting are averaged before ranking. Writing-style traits and composite indices are excluded. This describes similarity, not an overall quality ranking.

Filter comparisons to this model →

Reports featuring this model

DeepSeek V4.1 Flash 1017 · DeepSeek →

Sources

Every plotted observation links to its evaluation. Reported configurations, benchmark versions and task subsets can differ; similarity describes shared scores, not training lift.

  1. ARC Prize - Leaderboard ↗

    ARC Prize Foundation

  2. Arena leaderboard — WebDev evaluation ↗

    Arena

  3. AA-3D: Interactive 3D Experience Benchmark | Artificial Analysis ↗

    Artificial Analysis

  4. APEX-Agents-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  5. AA-LCR v1.1: Long Context Reasoning Benchmark | Artificial Analysis ↗

    Artificial Analysis

  6. CritPt v2 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  7. EnterpriseOps-Gym-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  8. GPQA Diamond Benchmark | Artificial Analysis ↗

    Artificial Analysis

  9. Harvey LAB-AA v1.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  10. HealthBench Professional Benchmark | Artificial Analysis ↗

    Artificial Analysis

  11. ITBench-AA Benchmark | Artificial Analysis ↗

    Artificial Analysis

  12. MMMU-Pro Benchmark | Artificial Analysis ↗

    Artificial Analysis

  13. SciCode Benchmark | Artificial Analysis ↗

    Artificial Analysis

  14. 𝜏³-Banking Benchmark | Artificial Analysis ↗

    Artificial Analysis

  15. Terminal-Bench 2.1 Benchmark | Artificial Analysis ↗

    Artificial Analysis

  16. DTBench — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  17. LMCA — Conceptual Reasoning Index ↗

    Conceptual Reasoning Index

  18. DeepSWE ↗

    Datacurve

  19. ARC-AGI-2 ↗

    Epoch AI

  20. Chess Puzzles ↗

    Epoch AI

  21. CritPt v1 ↗

    Epoch AI

  22. FrontierMath Erdős ↗

    Epoch AI

  23. FrontierMath Tier 4 (v2) ↗

    Epoch AI

  24. FrontierMath Tiers 1-3 (v2) ↗

    Epoch AI

  25. Furniture Assembly Benchmark ↗

    Epoch AI

  26. GPQA Diamond ↗

    Epoch AI

  27. Mystery Game Puzzles ↗

    Epoch AI

  28. OTIS Mock AIME 2024-2025 ↗

    Epoch AI

  29. SimpleQA Verified ↗

    Epoch AI

  30. AI benchmarks & capabilities ↗

    Epoch AI

  31. GDPx v1.0: The General Agentic Benchmark — SemiAnalysis ↗

    SemiAnalysis · 2026

  32. WeirdML ↗

    Håvard Tveit Ihle

  33. deepseek-ai/DeepSeek-V4.1-Flash-1017 · Hugging Face ↗

    DeepSeek AI

  34. Kimi K3: Open Frontier Intelligence — official model card ↗

    Moonshot AI · 2026

  35. DistressBench — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  36. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  37. HiL-Bench (Human-in-Loop Benchmark) — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  38. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  39. Humanity's Sixth Sense — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  40. Scale Labs — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  41. SWE-Bench Pro V2 — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  42. SWE-Bench Pro V2 — evaluation methodology and results ↗

    Scale Labs / SEAL · 2026

  43. AI Model Benchmarks — LM Council ↗

    LM Council

  44. GDP.pdf Benchmark | Surge AI ↗

    Surge AI

  45. FrontierSWE — software engineering benchmark ↗

    FrontierSWE

  46. ErdősBench: research mathematics, scoped progress and proof audits ↗

    Ulam AI · 2026

  47. BioMysteryBench — evaluation methodology and results ↗

    Vals AI · 2026

  48. Code Migration v1 — evaluation methodology and results ↗

    Vals AI · 2026

  49. Code Migration — evaluation methodology and results ↗

    Vals AI · 2026

  50. CorpFin — evaluation methodology and results ↗

    Vals AI · 2026

  51. CyberBench — evaluation methodology and results ↗

    Vals AI · 2026

  52. Excel Modeling Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  53. Excel Modeling Benchmark — evaluation methodology and results ↗

    Vals AI · 2026

  54. Finance Agent v2 — evaluation methodology and results ↗

    Vals AI · 2026

  55. Finance Agent — evaluation methodology and results ↗

    Vals AI · 2026

  56. Vals Finance Index — evaluation methodology and results ↗

    Vals AI · 2026

  57. GPQA Diamond — evaluation methodology and results ↗

    Vals AI · 2026

  58. HLAB-Vals-1.1 — evaluation methodology and results ↗

    Vals AI · 2026

  59. Harvey's Legal Agent Benchmark v1 — evaluation methodology and results ↗

    Vals AI · 2026

  60. IOI — evaluation methodology and results ↗

    Vals AI · 2026

  61. LiveCodeBench — evaluation methodology and results ↗

    Vals AI · 2026

  62. LegalBench — evaluation methodology and results ↗

    Vals AI · 2026

  63. Vals Legal Index — evaluation methodology and results ↗

    Vals AI · 2026

  64. Legal Research Bench — evaluation methodology and results ↗

    Vals AI · 2026

  65. MedCode — evaluation methodology and results ↗

    Vals AI · 2026

  66. MedScribe — evaluation methodology and results ↗

    Vals AI · 2026

  67. SAFE-Teen: Teen Mental Health & Safety | Vals AI A conversation branches after the chatbot turns down a request — evaluation methodology and results ↗

    Vals AI · 2026

  68. MMLU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  69. MMMU Pro — evaluation methodology and results ↗

    Vals AI · 2026

  70. MortgageTax — evaluation methodology and results ↗

    Vals AI · 2026

  71. ProgramBench — evaluation methodology and results ↗

    Vals AI · 2026

  72. ProofBench — evaluation methodology and results ↗

    Vals AI · 2026

  73. Public Benefits Bench — evaluation methodology and results ↗

    Vals AI · 2026

  74. Vals RSI Index — evaluation methodology and results ↗

    Vals AI · 2026

  75. SAGE — evaluation methodology and results ↗

    Vals AI · 2026

  76. SWE-bench Verified — evaluation methodology and results ↗

    Vals AI · 2026

  77. SWE-bench Multimodal — evaluation methodology and results ↗

    Vals AI · 2026

  78. Tax Agent Bench — evaluation methodology and results ↗

    Vals AI · 2026

  79. TaxEval — evaluation methodology and results ↗

    Vals AI · 2026

  80. Terminal-Bench 4.0 — evaluation methodology and results ↗

    Vals AI · 2026

  81. Terminal-Bench-Science — evaluation methodology and results ↗

    Vals AI · 2026

  82. Time Horizon Index: KSP — evaluation methodology and results ↗

    Vals AI · 2026

  83. Vals Index — evaluation methodology and results ↗

    Vals AI · 2026

  84. Vals Multimodal Index — evaluation methodology and results ↗

    Vals AI · 2026

  85. Vibe Code Bench 1-100 — evaluation methodology and results ↗

    Vals AI · 2026

  86. Vibe Code Bench — evaluation methodology and results ↗

    Vals AI · 2026

  87. Surface Evolver Bench ↗

    yhenon

Names used in source tables

K3 · Kimi K3 · Kimi K3 (Max) · Kimi K3 (Mini-SWE-agent) · Kimi K3 (OpenHands) · Kimi K3 (high) · Kimi K3 (low) · Kimi K3 (max) · Kimi K3 max · Kimi k3 · Kimi-K3 (max) · Kimi-K3 (mini-swe-agent) max · kimi-k3 · kimi-k3 (max) · kimi-k3 max