Which model leads on Chess Puzzles?

Claude Fable 5.2 from Anthropic leads Chess Puzzles with a score of 77.9. The median score across 104 tested models is 21.6.

Is Chess Puzzles saturated?

No · the top score is 77.9 out of 100 (78%). There is still meaningful room for improvement on Chess Puzzles.

Does Chess Puzzles predict performance on other benchmarks?

Yes · Chess Puzzles scores correlate 0.88 with Exploitbench across 7 shared models. Models that do well on Chess Puzzles tend to do well on Exploitbench.

How often is Chess Puzzles data refreshed?

BenchGecko pulls updates daily. New model scores on Chess Puzzles appear as soon as they are published by Epoch AI or the model provider.

Benchmark · ReasoningCompetitive

Chess Puzzles

Name: Chess Puzzles Benchmark
Creator: BenchGecko
License: https://creativecommons.org/licenses/by/4.0/

Chess Puzzles · tests strategic and tactical reasoning by having models solve chess puzzle positions, evaluating lookahead and pattern recognition abilities.

Updated 2026-11-09

Models tested

104

Top score

77.9

Claude Fable 5.2

Median

21.6

min 0.0

Top-5 spread

σ 5.2

wide open

The Frontier

Best score over time · one chart, every benchmark

Chart type

Frontier on Chess Puzzles rose from 4.3 to 77.9 in 24 months · +73.7 points · latest leader Claude Fable 5.2 from Anthropic.

Pink dots = frontier records · 13 totalClick to open model page

Full rankings

104 models tested · sorted by score

#	Model	Score	Price
1	Claude Fable 5.2· Anthropic	77.9	$8.00
2	GPT-6 Astra· OpenAI	74.8	$10.00
3	GPT-6.1 Sol· OpenAI	66.3	$2.00
4	GPT-5.6 Sol· OpenAI	66.3	$4.00
5	Claude Opus 5.5· Anthropic	65.3	$4.00
6	Gemini 3.7 Flash· Google DeepMind	64.2	$0.75
7	GPT-5.5· OpenAI	61.1	$5.00
8	GPT-5.5 Pro· OpenAI	60.0	$30.00
9	GPT-5.4 Pro· OpenAI	58.6	$30.00
10	GPT-5.2 Pro· OpenAI	57.9	$21.00
11	Gemini 4 Argon· Google DeepMind	56.9	$2.00
12	Claude Sonnet 5.5· Anthropic	56.9	$2.00
13	Gemini 3.8 Flash· Google DeepMind	55.8	$0.82
14	GPT-6 Sol· OpenAI	53.7	$2.00
15	Claude Haiku 5.5· Anthropic	52.6	$0.10
16	Claude Opus 4.8· Anthropic	49.5	$5.00
17	Gemini 3.1 Pro Preview· Google DeepMind	48.4	$2.00
18	GPT-5 Pro· OpenAI	45.3	$15.00
19	GPT-5.2· OpenAI	45.3	$1.75
20	Claude Fable 5.1· Anthropic	45.3	$10.00
21	GPT-5.4· OpenAI	45.3	$2.50
22	DeepSeek V4.1 Flash· DeepSeek	45.3	$0.05
23	Claude Fable 5· Anthropic	44.2	$10.00
24	GPT-5.6 Terra· OpenAI	44.2	$2.00
25	Qwen3.8 Max (0902)· Alibaba Qwen	43.2	$2.00
26	DeepSeek V4.1 Flash 1017· DeepSeek	43.2	$0.01
27	DeepSeek V4 Pro 0813· DeepSeek	43.2	$0.11
28	GPT-5.4 Nano· OpenAI	42.1	$0.20
29	Gemini 3.5 Flash· Google DeepMind	42.1	$1.50
30	Kimi K3.1 Pro· moonshotai	42.1	$2.00
31	Claude Sonnet 5· Anthropic	42.1	$2.00
32	GPT-5.6 Luna· OpenAI	40.0	$0.20
33	Claude Opus 5· Anthropic	40.0	$5.00
34	GPT-5.1· OpenAI	39.0	$1.25
35	Gemini 3.6 Flash· Google DeepMind	39.0	$0.75
36	Gemini 3 Flash Preview· Google DeepMind	38.0	$0.50
37	Claude Opus 4.7· Anthropic	37.9	$5.00
38	Grok 4.7· xAI	37.9	$2.00
39	GPT-5· OpenAI	37.0	$1.25
40	GLM 5.3· z-ai	33.7	$0.07
41	Kimi K3· moonshotai	33.7	$0.19
42	o3· OpenAI	33.7	$2.00
43	GPT-6 Luna· OpenAI	32.7	$0.10
44	Grok 4.5· xAI	31.6	$2.00
45	Grok 4.6· xAI	29.5	$2.00
46	DeepSeek V4 Flash 0731· DeepSeek	28.4	$0.04
47	Grok 4· xAI	28.0	n/a
48	Gemini 3 Pro· Google DeepMind	27.4	n/a
49	Grok 4.20· xAI	25.3	$1.25
50	Mistral Large 4· Mistral AI	25.3	$0.70
51	Qwen3.7 Max· Alibaba Qwen	23.2	$1.48
52	Qwen3.6 35B A3B· Alibaba Qwen	22.1	$0.10
53	Gemini 3.1 Flash Lite· Google DeepMind	21.1	$0.25
54	Qwen3.7 Plus· Alibaba Qwen	20.0	$0.32
55	GPT-5 Mini· OpenAI	20.0	$0.25
56	Kimi K2 Thinking· moonshotai	20.0	$0.60
57	Qwen3.7 Flash· Alibaba Qwen	19.0	$0.03
58	GPT-5.4 Mini· OpenAI	19.0	$0.75
59	GLM 5.2· z-ai	17.9	$0.01
60	o4 Mini· OpenAI	17.9	$1.10
61	Qwen3.6 27B· Alibaba Qwen	17.9	$0.32
62	o4 Mini (batch)· OpenAI	17.9	$0.55
63	GPT-5 Nano· OpenAI	17.9	$0.05
64	Claude Opus 4.6· Anthropic	17.0	$5.00
65	GLM 4.7· z-ai	16.9	$0.60
66	Gemini 3.5 Flash Lite· Google DeepMind	16.9	$0.30
67	GLM 5.3 Flash· z-ai	16.9	$0.04
68	gpt-oss-120b· OpenAI	15.8	$0.15
69	Kimi K2.6· moonshotai	15.8	$0.65
70	Gemini 2.5 Pro· Google DeepMind	15.8	$1.25
71	Qwen3.6 Max Preview· Alibaba Qwen	15.8	$1.03
72	GLM 5.1· z-ai	15.8	$1.33
73	DeepSeek V4 Pro· DeepSeek	14.8	$0.88
74	Grok 4.3· xAI	13.7	$1.25
75	Claude Sonnet 4.6· Anthropic	13.0	$3.00
76	MiniMax M3· minimax	12.7	$0.30
77	Claude Opus 4.5· Anthropic	12.0	$5.00
78	Kimi K2.5· moonshotai	12.0	$0.53
79	Qwen3.6 Plus· Alibaba Qwen	11.6	$0.33
80	o3 Mini (batch)· OpenAI	10.6	$0.55
81	o3 Mini· OpenAI	10.6	$1.10
82	GLM 5· z-ai	10.0	$1.00
83	DeepSeek V3.2· DeepSeek	9.5	$0.26
84	Kimi K2.7 Code· moonshotai	9.5	$0.60
85	Qwen3.5-Flash· Alibaba Qwen	9.5	$0.07
86	Qwen3.5 397B A17B· Alibaba Qwen	8.5	$0.60
87	Qwen3.6 Flash· Alibaba Qwen	7.4	$0.19
88	Nemotron 3 Ultra· NVIDIA	7.4	$0.60
89	Qwen3.5-9B· Alibaba Qwen	7.4	$0.17
90	Qwen3 235B A22B Thinking 2507· Alibaba Qwen	7.4	$0.45
91	o1· OpenAI	6.4	$15.00
92	GPT-4.1· OpenAI	6.4	$2.00
93	Claude Sonnet 4.5· Anthropic	5.3	$3.00
94	Qwen3.5-35B-A3B· Alibaba Qwen	5.3	$0.14
95	GPT-4o (2024-11-20)· OpenAI	4.3	$2.50
96	Qwen 3.5 Plus (hosted 397B-A17B)· Alibaba Qwen	4.3	n/a
97	Qwen3 Max· Alibaba Qwen	4.0	$0.78
98	Claude Haiku 4.5· Anthropic	3.2	$1.00
99	Claude Opus 4.1· Anthropic	2.1	$15.00
100	Gemma 4 26B A4B · Google DeepMind	1.1	$0.08
101	Qwen3 8B· Alibaba Qwen	0.0	$0.12
102	QwQ 32B· Alibaba Qwen	0.0	$0.15
103	Gemma 4 31B· Google DeepMind	0.0	$0.20
104	Qwen3 32B· Alibaba Qwen	0.0	$0.08