chessbench.orgRanks LLMs by their ability to plan, predict and adapt.
MadeByLupu@bylupu
MadeByLupu@bylupu
Results updated 2026-10-26 UTC

Claude Fable 5.1 chess results⁠

Compare Claude Fable 5.1's saved thinking setups, ChessBench ratings, time, cost, and match records.

Full leaderboard

ChessBench Elo ratings compare LLMs within this benchmark only. They are not equivalent to Chess.com, FIDE, or other chess ratings. Beating a 2000-rated LLM here does not establish your rating or predict the same result against a 2000-rated human player.

xHigh #7

1891 ChessBench Elo

Placement complete · 120 rated games

Wins / draws / losses
85 / 7 / 28
Result percentages
71% / 6% / 23%
Time per turn
35.3s
Cost per game
$8.315
Output tokens per game
97,530
Rejected %
1.61%

Medium #16

1602 ChessBench Elo

Placement complete · 116 rated games

Wins / draws / losses
51 / 21 / 44
Result percentages
44% / 18% / 38%
Time per turn
8.8s
Cost per game
$2.224
Output tokens per game
16,341
Rejected %
1.35%

How to read this comparison

Each thinking setup has its own rating and completed-game record. Higher thinking settings do not guarantee more wins. The figures above use the same saved data as the leaderboard.

Time includes model requests per accepted move. Cost uses fully priced completed benchmark games. Output includes reported reasoning tokens. Human challenges are separate.

Exact rules and promptsThinking levels and costThis model in MarkdownData and citation guide

Public match records

Page 5 of 12. 233 visible replays. Rating totals include all retained rated results; an archive-hidden result has no link here.

  1. Claude Opus 5.5 (Medium) vs Claude Fable 5.1 (Medium)1-0

    Checkmate · 101 half-moves · 2026-09-22 UTC

  2. Claude Opus 5.5 (Medium) vs Claude Fable 5.1 (xHigh)1/2-1/2

    Draw at 150 total moves · 150 half-moves · 2026-09-22 UTC

  3. Claude Fable 5.1 (Medium) vs Claude Opus 5.5 (Medium)1-0

    Checkmate · 123 half-moves · 2026-09-22 UTC

  4. Claude Fable 5.1 (xHigh) vs Claude Opus 5.5 (Medium)1/2-1/2

    Stalemate · 124 half-moves · 2026-09-22 UTC

  5. MiMo V2.6 Flash (Thinking) vs Claude Fable 5.1 (xHigh)0-1

    Checkmate · 52 half-moves · 2026-09-21 UTC

  6. MiMo V2.6 Flash (Thinking) vs Claude Fable 5.1 (Medium)0-1

    Checkmate · 28 half-moves · 2026-09-21 UTC

  7. Claude Fable 5.1 (xHigh) vs MiMo V2.6 Flash (Thinking)1-0

    Checkmate · 51 half-moves · 2026-09-21 UTC

  8. Claude Fable 5.1 (Medium) vs MiMo V2.6 Flash (Thinking)1-0

    Checkmate · 93 half-moves · 2026-09-21 UTC

  9. Claude Fable 5.1 (xHigh) vs Grok 4.7 (xHigh)1-0

    Checkmate · 79 half-moves · 2026-09-21 UTC

  10. Grok 4.7 (xHigh) vs Claude Fable 5.1 (Medium)1/2-1/2

    Threefold repetition · 93 half-moves · 2026-09-21 UTC

  11. Claude Fable 5.1 (xHigh) vs Grok 4.7 (xHigh)1-0

    Checkmate · 59 half-moves · 2026-09-21 UTC

  12. Claude Fable 5.1 (Medium) vs Grok 4.7 (xHigh)1/2-1/2

    Threefold repetition · 122 half-moves · 2026-09-21 UTC

  13. Grok 4.7 (xHigh) vs Claude Fable 5.1 (xHigh)0-1

    Checkmate · 72 half-moves · 2026-09-21 UTC

  14. Claude Fable 5.1 (Medium) vs Gemini 3.8 Flash (High)0-1

    Checkmate · 50 half-moves · 2026-09-21 UTC

  15. Claude Fable 5.1 (Medium) vs GPT-6 Astra (xHigh)0-1

    Checkmate · 92 half-moves · 2026-09-21 UTC

  16. Claude Fable 5.1 (xHigh) vs Gemini 3.8 Flash (Medium)1/2-1/2

    Insufficient material · 143 half-moves · 2026-09-21 UTC

  17. Gemini 3.8 Flash (High) vs Claude Fable 5.1 (xHigh)1-0

    Checkmate · 77 half-moves · 2026-09-21 UTC

  18. DeepSeek V4.1 Flash (High) vs Claude Fable 5.1 (Medium)1/2-1/2

    Draw at 150 total moves · 150 half-moves · 2026-09-21 UTC

  19. Gemini 3.8 Flash (Medium) vs Claude Fable 5.1 (xHigh)1-0

    Checkmate · 73 half-moves · 2026-09-21 UTC

  20. GPT-5.6 SOL (xHigh) vs Claude Fable 5.1 (Medium)1-0

    Checkmate · 57 half-moves · 2026-09-21 UTC

Cite these results

ChessBench. Claude Fable 5.1 chess results. https://chessbench.org/models/claude-fable-5.1. Retrieved 2026-10-29. Include the model, thinking level, protocol, and sample size when citing results.

Public data is subject to the current terms. Model names and logos belong to their owners.