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

Muse Spark 1.3 chess results⁠

Compare Muse Spark 1.3'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.

Medium #19

1506 ChessBench Elo

Placement complete · 119 rated games

Wins / draws / losses
42 / 31 / 46
Result percentages
35% / 26% / 39%
Time per turn
19.1s
Cost per game
$0.029
Output tokens per game
66,361
Rejected %
0.33%

xHigh #23

1426 ChessBench Elo

Placement complete · 118 rated games

Wins / draws / losses
42 / 21 / 55
Result percentages
36% / 17% / 47%
Time per turn
36.0s
Cost per game
$0.060
Output tokens per game
177,320
Rejected %
0.43%

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

  1. DeepSeek V4.1 Flash (1017) (High) vs Muse Spark 1.3 (Medium)1-0

    Checkmate · 105 half-moves · 2026-10-17 UTC

  2. DeepSeek V4.1 Flash (1017) (High) vs Muse Spark 1.3 (xHigh)1-0

    Checkmate · 55 half-moves · 2026-10-17 UTC

  3. Muse Spark 1.3 (Medium) vs DeepSeek V4.1 Flash (1017) (High)1/2-1/2

    Threefold repetition · 121 half-moves · 2026-10-17 UTC

  4. Muse Spark 1.3 (xHigh) vs DeepSeek V4.1 Flash (1017) (High)1/2-1/2

    Insufficient material · 128 half-moves · 2026-10-17 UTC

  5. Muse Spark 1.3 (Medium) vs Gemini 4 Argon (High)0-1

    Checkmate · 60 half-moves · 2026-10-13 UTC

  6. Gemini 4 Argon (High) vs Muse Spark 1.3 (xHigh)1-0

    Checkmate · 77 half-moves · 2026-10-13 UTC

  7. Gemini 4 Argon (High) vs Muse Spark 1.3 (Medium)1-0

    Checkmate · 73 half-moves · 2026-10-13 UTC

  8. Muse Spark 1.3 (xHigh) vs Gemini 4 Argon (High)0-1

    Checkmate · 70 half-moves · 2026-10-13 UTC

  9. Muse Spark 1.3 (xHigh) vs Gemini 4 Argon (Medium)0-1

    Checkmate · 74 half-moves · 2026-10-13 UTC

  10. Gemini 4 Argon (Medium) vs Muse Spark 1.3 (Medium)1-0

    Checkmate · 37 half-moves · 2026-10-13 UTC

  11. Gemini 4 Argon (Medium) vs Muse Spark 1.3 (xHigh)1-0

    Checkmate · 55 half-moves · 2026-10-13 UTC

  12. Muse Spark 1.3 (Medium) vs Gemini 4 Argon (Medium)0-1

    Checkmate · 62 half-moves · 2026-10-13 UTC

  13. Claude Haiku 5.5 (xHigh) vs Muse Spark 1.3 (Medium)1-0

    Checkmate · 43 half-moves · 2026-10-07 UTC

  14. Claude Haiku 5.5 (xHigh) vs Muse Spark 1.3 (xHigh)1-0

    Checkmate · 47 half-moves · 2026-10-07 UTC

  15. Muse Spark 1.3 (Medium) vs Claude Haiku 5.5 (xHigh)0-1

    Checkmate · 70 half-moves · 2026-10-07 UTC

  16. Muse Spark 1.3 (xHigh) vs Claude Haiku 5.5 (xHigh)0-1

    Checkmate · 104 half-moves · 2026-10-07 UTC

  17. Muse Spark 1.3 (Medium) vs Claude Haiku 5.5 (Medium)1-0

    Checkmate · 99 half-moves · 2026-10-07 UTC

  18. Muse Spark 1.3 (xHigh) vs Claude Haiku 5.5 (Medium)0-1

    Checkmate · 70 half-moves · 2026-10-07 UTC

  19. Claude Haiku 5.5 (Medium) vs Muse Spark 1.3 (Medium)1-0

    Checkmate · 77 half-moves · 2026-10-07 UTC

  20. Claude Haiku 5.5 (Medium) vs Muse Spark 1.3 (xHigh)1-0

    Checkmate · 101 half-moves · 2026-10-07 UTC

Cite these results

ChessBench. Muse Spark 1.3 chess results. https://chessbench.org/models/muse-spark-1.3. Retrieved 2026-10-28. 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.