modelgrep

Smartest LLMs

Quick answer · Updated June 2026

The smartest LLM is Claude Fable 5, scoring 64.9 on the Artificial Analysis Intelligence Index. Claude Opus 4.8 (61.4) and Claude Opus 4.7 (57.3) round out the top three.

64.9Intelligence
$10.00Input /M
1MContext

The most capable AI models ranked by the Artificial Analysis Intelligence Index — a composite of reasoning, knowledge, math and science benchmarks. These are the highest-scoring large language models available today.

  1. 1A
    claude-fable-5
    ReasoningToolsJSON+1$10.00/M · 1M ctx
    64.9
    Intelligence
  2. 2A
    claude-opus-4.8
    ReasoningToolsJSON+1$5.00/M · 58 t/s · 1.7s ttft
    61.4
    Intelligence
  3. 3A
    claude-opus-4.7
    ReasoningToolsJSON+1$5.00/M · 60 t/s · 1.5s ttft
    57.3
    Intelligence
  4. 4O
    gpt-5.5
    ReasoningToolsJSON+1$5.00/M · 37 t/s · 3.3s ttft
    56.7
    Intelligence
  5. 5Q
    qwen3.7-max
    ReasoningToolsJSON$1.25/M · 39 t/s · 1.4s ttft
    56.6
    Intelligence
  6. 6M
    minimax-m3
    ReasoningToolsJSON+1$0.300/M · 47 t/s · 720ms ttft
    54.7
    Intelligence
  7. 7X
    mimo-v2.5-pro
    ReasoningToolsJSON$0.435/M · 32 t/s · 312ms ttft
    53.8
    Intelligence
  8. 8O
    gpt-5.3-codex
    ReasoningToolsJSON+1$1.75/M · 45 t/s · 4.0s ttft
    53.6
    Intelligence
  9. 9Q
    qwen3.7-plus
    ReasoningToolsJSON+1$0.320/M · 29 t/s · 1.0s ttft
    53.3
    Intelligence
  10. 10X
    grok-4.3
    ReasoningToolsJSON+1$1.25/M · 135 t/s · 772ms ttft
    53.2
    Intelligence
  11. 11A
    claude-opus-4.6
    ReasoningToolsJSON+1$5.00/M · 46 t/s · 1.2s ttft
    52.9
    Intelligence
  12. 12Q
    qwen3.6-max-preview
    ReasoningToolsJSON$1.04/M · 48 t/s · 1.4s ttft
    51.8
    Intelligence
  13. 13Q
    qwen3.6-plus
    ReasoningToolsJSON+1$0.325/M · 36 t/s · 1.4s ttft
    50.0
    Intelligence
  14. 14Z
    glm-5
    ReasoningToolsJSON$0.600/M · 112 t/s · 189ms ttft
    49.8
    Intelligence
  15. 15M
    minimax-m2.7
    ReasoningToolsJSON$0.250/M · 334 t/s · 673ms ttft
    49.6
    Intelligence
  16. 16X
    mimo-v2.5
    ReasoningToolsJSON+2$0.140/M · 49 t/s · 3.0s ttft
    49.0
    Intelligence
  17. 17O
    gpt-5.2-codex
    ReasoningToolsJSON+1$1.75/M · 85 t/s · 2.4s ttft
    49.0
    Intelligence
  18. 18N
    nemotron-3-ultra-550b-a55b:free
    ReasoningToolsFree/M · 64 t/s · 1.2s ttft
    47.7
    Intelligence
  19. 19N
    nemotron-3-ultra-550b-a55b
    ReasoningToolsJSON$0.500/M · 64 t/s · 1.2s ttft
    47.7
    Intelligence
  20. 20O
    gpt-5.1
    ReasoningToolsJSON+1$1.25/M · 55 t/s · 1.1s ttft
    47.7
    Intelligence
  21. 21Z
    glm-5-turbo
    ReasoningToolsJSON$1.20/M · 15 t/s · 2.0s ttft
    46.8
    Intelligence
  22. 22G
    gemini-3-flash-preview
    ReasoningToolsJSON+2$0.500/M · 66 t/s · 1.0s ttft
    46.4
    Intelligence
  23. 23D
    deepseek-v4-flash
    ReasoningToolsJSON$0.098/M · 79 t/s · 582ms ttft
    46.0
    Intelligence
  24. 24O
    gpt-5-codex
    ReasoningToolsJSON+1$1.25/M · 43 t/s · 3.4s ttft
    44.6
    Intelligence
  25. 25O
    gpt-5.4-nano
    ReasoningToolsJSON+1$0.200/M · 56 t/s · 626ms ttft
    44.0
    Intelligence

Frequently asked

What is the smartest LLM?

The smartest LLM is Claude Fable 5, scoring 64.9 on the Artificial Analysis Intelligence Index. Claude Opus 4.8 (61.4) and Claude Opus 4.7 (57.3) round out the top three.

What's a good alternative to Claude Fable 5?

Claude Opus 4.8 (61.4) is the closest alternative on this metric, followed by Claude Opus 4.7 (57.3). See the full ranking above for the tradeoffs.

By maker

All rankings