gpt-5.4-mini vs gpt-5.5

Season 1 · Round 2 · 31 July 2026

Seat	Model	Score
gpt-5.4-mini	openai/gpt-5.4-mini	0
gpt-5.5	openai/gpt-5.5	1

Won by gpt-5.5 on elimination · 42 turns · $3.27 of model time

Move by move

This match was played before reasoning capture. Where a model wrote out its analysis in its reply, it is shown as written analysis. Hidden thinking was not recorded.

gpt-5.5 thought for 541 s as White · gpt-5.4-mini thought for 16 s as Black

1. e4 gpt-5.5 5 s best

Thought for 5 s. The provider returned no reasoning text for this move.

1... e5 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

2. Nf3 gpt-5.5 3 s best

Thought for 3 s. The provider returned no reasoning text for this move.

2... Nc6 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

3. Bb5 gpt-5.5 8 s best

Thought for 8 s. The provider returned no reasoning text for this move.

3... Nf6 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

4. O-O gpt-5.5 10 s best

Thought for 10 s. The provider returned no reasoning text for this move.

4... Nxe4 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

5. d4 gpt-5.5 10 s best

Thought for 10 s. The provider returned no reasoning text for this move.

5... Nxd4 gpt-5.4-mini 1 s blunder

Played from memory: no reasoning was produced for this move.

6. Nxd4 gpt-5.5 36 s best

Thought for 36 s. The provider returned no reasoning text for this move.

6... Nxf2 gpt-5.4-mini 1 s mistake

Played from memory: no reasoning was produced for this move.

7. Rxf2 gpt-5.5 57 s good

Thought for 57 s. The provider returned no reasoning text for this move.

7... e4 gpt-5.4-mini 1 s mistake

Played from memory: no reasoning was produced for this move.

8. Nc6 gpt-5.5 11 s blunder

Thought for 11 s. The provider returned no reasoning text for this move.

8... d5 gpt-5.4-mini 1 s blunder

Played from memory: no reasoning was produced for this move.

9. Nxd8+ gpt-5.5 21 s best

Thought for 21 s. The provider returned no reasoning text for this move.

9... Kxd8 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

10. Qxd5+ gpt-5.5 31 s best

Thought for 31 s. The provider returned no reasoning text for this move.

10... Bd6 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

11. Be2 gpt-5.5 79 s good forced substitution

Thought for 79 s. The provider returned no reasoning text for this move.

11... Bd7 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

12. Qxb7 gpt-5.5 69 s best

Thought for 69 s. The provider returned no reasoning text for this move.

12... Bxh2+ gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

13. Kxh2 gpt-5.5 11 s best

Thought for 11 s. The provider returned no reasoning text for this move.

13... Bf5 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

14. Qxa8+ gpt-5.5 33 s good

Thought for 33 s. The provider returned no reasoning text for this move.

14... Kd7 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

15. Qxh8 gpt-5.5 29 s good

Thought for 29 s. The provider returned no reasoning text for this move.

15... e3 gpt-5.4-mini 1 s good

Played from memory: no reasoning was produced for this move.

16. Rxf5 gpt-5.5 47 s best

Thought for 47 s. The provider returned no reasoning text for this move.

16... Ke7 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

17. Bxe3 gpt-5.5 66 s best

Thought for 66 s. The provider returned no reasoning text for this move.

17... Kd6 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

18. Qd8+ gpt-5.5 11 s best

Thought for 11 s. The provider returned no reasoning text for this move.

18... Kc6 gpt-5.4-mini 1 s best

Played from memory: no reasoning was produced for this move.

19. Qd5# gpt-5.5 6 s brilliant

Thought for 6 s. The provider returned no reasoning text for this move.

What this match ran under

Tokens per move: 4096
Mode: assisted
Hint after an illegal move: yes
When it keeps failing: random legal

Why these exist, and what removing them costs.