Noometry

Reasoning benchmark

Chess Puzzles leaderboard

As of October 2026, GPT-6 Astra has the highest published Chess Puzzles score on Noometry at 72%, out of 129 models with results.

Last verified

About Chess Puzzles

Chess tactics puzzles solved from a text position, without an engine.

Category
Reasoning
Introduced
2025
Format
Move prediction
Unit
Percent (random guessing ≈ 5%)
Official site
epoch.ai

Top 15 models

  1. GPT-6 Astra 72%
  2. GPT-5.5 Pro 64%
  3. GPT-5.6 Sol 64%
  4. Gemini 3.8 Flash 61%
  5. GPT-6.1 Sol 61%
  6. GPT-5.4 Pro 58.6%
  7. Gemini 3.1 Pro Preview 55%
  8. GPT-5.5 54%
  9. GPT-5.6 Terra 54%
  10. Gemini 3.5 Flash 50%
  11. GPT-5.2 49%
  12. Claude Fable 5.1 47%
  13. DeepSeek V4 Pro 47%
  14. Gemini 3.7 Flash 47%
  15. GPT-5.4 44%

Sponsored placements are available on pages like this one. Advertise on Noometry

All results

#ModelProviderScoreSettingSourceDate
1GPT-6 Astra OpenAI72%maxEpoch AI2026-08-30
2GPT-5.5 Pro OpenAI64%xhighEpoch AI2026-05-01
3GPT-5.6 Sol OpenAI64%promaxEpoch AI2026-07-10
4Gemini 3.8 Flash Google61%highEpoch AI2026-09-02
5GPT-6.1 Sol OpenAI61%maxEpoch AI2026-09-29
6GPT-5.4 Pro OpenAI58.6%xhighEpoch AI2026-03-19
7Gemini 3.1 Pro Preview Google55%Epoch AI2026-02-19
8GPT-5.5 OpenAI54%xhighEpoch AI2026-04-24
9GPT-5.6 Terra OpenAI54%maxEpoch AI2026-07-09
10Gemini 3.5 Flash Google50%highEpoch AI2026-05-28
11GPT-5.2 OpenAI49%xhighEpoch AI2025-12-15
12Claude Fable 5.1 Anthropic47%maxEpoch AI2026-09-01
13DeepSeek V4 Pro DeepSeek47%maxEpoch AI2026-08-18
14Gemini 3.7 Flash Google47%highEpoch AI2026-08-14
15GPT-5.4 OpenAI44%xhighEpoch AI2026-03-11
16Gemini 3.6 Flash Google43%lowEpoch AI2026-08-07
17Claude Opus 5 Anthropic42%maxEpoch AI2026-07-24
18Claude Fable 5 Anthropic41%highEpoch AI2026-08-06
19Gemini 3 Flash Preview Google40%highEpoch AI2026-08-06
20GPT-5.6 Luna OpenAI40%maxEpoch AI2026-07-09
21Grok 4.6 xAI40%highEpoch AI2026-08-12
22Qwen3.8 Max Alibaba (Qwen)40%xhighEpoch AI2026-09-02
23Kimi K3 Moonshot AI39%maxEpoch AI2026-07-16
24Grok 4.7 xAI38%xhighEpoch AI2026-09-22
25Muse Spark 1.3 Meta38%maxEpoch AI2026-09-18
26o3 OpenAI38%mediumEpoch AI2026-08-07
27GPT-5 OpenAI37%highEpoch AI2025-12-08
28Grok 4.5 xAI36%highEpoch AI2026-07-08
29Claude Sonnet 5 Anthropic35%xhighEpoch AI2026-06-30
30Claude Opus 4.8 Anthropic34%maxEpoch AI2026-05-29
31DeepSeek V4 Flash DeepSeek33%maxEpoch AI2026-08-02
32GPT-5.1 OpenAI32%highEpoch AI2025-12-08
33Gemini 3 Pro Google31%Epoch AI2025-12-08
34GPT-6 Luna OpenAI31%maxEpoch AI2026-09-22
35Claude Opus 4.7 Anthropic30%xhighEpoch AI2026-04-20
36GPT-5.4 nano OpenAI30%highEpoch AI2026-04-14
37GPT-5 Mini OpenAI30%highEpoch AI2026-08-07
38Grok 4 xAI28%Epoch AI2026-01-30
39GPT-5 Nano OpenAI27%highEpoch AI2026-08-07
40Kimi K2.6 Moonshot AI26%Epoch AI2026-05-07
41o4-mini OpenAI26%highEpoch AI2025-12-08
42Qwen3.6 35B-A3B Alibaba (Qwen)26%Epoch AI2026-08-07
43Gemini 3.1 Flash Lite Google25%lowEpoch AI2026-08-06
44Grok 4.3 xAI25%highEpoch AI2026-06-17
45GPT-5.4 mini OpenAI24%xhighEpoch AI2026-08-07
46Grok 4.20 (Non-Reasoning) xAI24%Epoch AI2026-07-13
47Qwen3.7 Plus Alibaba (Qwen)24%Epoch AI2026-08-07
48Qwen3.7 Flash Alibaba (Qwen)23%Epoch AI2026-08-07
49Gemini 3.5 Flash Lite Google22%highEpoch AI2026-08-06
50Qwen3.5 Plus Alibaba (Qwen)22%Epoch AI2026-08-07
51Qwen3.6 27B Alibaba (Qwen)22%Epoch AI2026-08-07
52GLM-5.2 Z.ai (Zhipu)21%maxEpoch AI2026-06-17
53GLM-5.3 Z.ai (Zhipu)21%maxEpoch AI2026-08-24
54Inkling Thinking Machines Lab21%xhighEpoch AI2026-08-05
55Kimi K2.7 Code Moonshot AI21%Epoch AI2026-08-07
56Qwen3.5-Flash Alibaba (Qwen)21%Epoch AI2026-08-07
57Gemini 2.5 Pro Google20%Epoch AI2025-12-08
58gpt-oss-120b OpenAI20%highEpoch AI2025-12-11
59Kimi K2 Thinking Turbo Moonshot AI20%Epoch AI2025-12-10
60Qwen3.6 Flash Alibaba (Qwen)20%Epoch AI2026-08-07
61Qwen3.6 Max Preview Alibaba (Qwen)20%Epoch AI2026-08-07
62GLM-5.1 Z.ai (Zhipu)19%Epoch AI2026-08-10
63Qwen3.7 Max Alibaba (Qwen)19%Epoch AI2026-08-07
64Inkling-Small Thinking Machines Lab18%xhighEpoch AI2026-08-14
65Claude Opus 4.6 Anthropic17%32KEpoch AI2026-02-06
66o3-mini OpenAI17%highEpoch AI2025-12-08
67Qwen3.6 Plus Alibaba (Qwen)17%Epoch AI2026-08-07
68o1 OpenAI15%highEpoch AI2026-08-07
69DeepSeek-V3.2-Exp DeepSeek14%Epoch AI2025-12-16
70GLM-5.3-Flash Z.ai (Zhipu)14%maxEpoch AI2026-08-26
71MiniMax-M3 MiniMax14%Epoch AI2026-08-10
72Claude Sonnet 4.6 Anthropic13%32KEpoch AI2026-02-20
73GPT-4o OpenAI13%Epoch AI2026-07-15
74Qwen3.5 397B-A17B Alibaba (Qwen)13%noneEpoch AI2026-08-07
75Claude Opus 4.5 Anthropic12%32KEpoch AI2025-12-08
76Claude Sonnet 4.5 Anthropic12%32KEpoch AI2025-12-08
77GPT-5.5 Instant OpenAI12%Epoch AI2026-08-02
78Kimi K2.5 Moonshot AI12%Epoch AI2026-01-28
79Nemotron 3 Ultra NVIDIA12%Epoch AI2026-08-10
80Qwen3 235B-A22B Alibaba (Qwen)12%Epoch AI2025-12-11
81Qwen3.5-9B Alibaba (Qwen)12%noneEpoch AI2026-08-27
82GLM-5 Z.ai (Zhipu)10%Epoch AI2026-02-12
83Qwen3.5 35B-A3B Alibaba (Qwen)10%noneEpoch AI2026-08-30
84Claude Haiku 4.5 Anthropic8%32KEpoch AI2026-07-16
85Qwen3-30B-A3B Alibaba (Qwen)8%Epoch AI2026-08-30
86Claude Opus 4.1 Anthropic7%Epoch AI2026-07-20
87GPT-4.1 mini OpenAI7%Epoch AI2026-07-15
88Gemma 4 26B A4B IT Google6%minimalEpoch AI2026-08-06
89GLM-4.7 Z.ai (Zhipu)6%Epoch AI2026-01-29
90GPT-4.1 OpenAI6%Epoch AI2026-08-07
91GPT-4 Turbo OpenAI6%Epoch AI2026-07-15
92Claude 3 Opus Anthropic5%Epoch AI2026-07-15
93Gemma 4 31B IT Google5%minimalEpoch AI2026-08-06
94Qwen3 32B Alibaba (Qwen)5%Epoch AI2026-08-28
95Qwen3 8B Alibaba (Qwen)5%Epoch AI2026-08-27
96QwQ-32B Alibaba (Qwen)5%Epoch AI2026-08-28
97GPT-4 OpenAI4%Epoch AI2026-08-07
98gpt-oss-20b OpenAI4%mediumEpoch AI2026-08-27
99Qwen3 14B Alibaba (Qwen)4%Epoch AI2026-08-30
100Qwen3-4B Alibaba (Qwen)4%Epoch AI2026-08-28
101Qwen3 Max Alibaba (Qwen)4%Epoch AI2025-12-10
102Magistral Small Mistral AI3%Epoch AI2026-08-30
103DeepSeek-R1-Distill-Qwen-14B DeepSeek1%Epoch AI2026-08-28
104DeepSeek-R1-Distill-Qwen-32B DeepSeek1%Epoch AI2026-08-28
105Mistral Small 3.1 Mistral AI1%Epoch AI2026-08-30
106Mistral Small 3.2 Mistral AI1%Epoch AI2026-08-30
107Phi-4 Microsoft1%Epoch AI2026-08-28
108DeepSeek LLM 67B DeepSeek0%Epoch AI2026-08-28
109DeepSeek-R1-Distill-Qwen-1.5B DeepSeek0%Epoch AI2026-08-28
110Gemma 3 12B Google0%Epoch AI2026-08-28
111Gemma 3 1B Google0%Epoch AI2026-08-28
112Gemma 3 27B Google0%Epoch AI2026-08-27
113Gemma 3 4B Google0%Epoch AI2026-08-28
114GLM-4.7-Flash Z.ai (Zhipu)0%Epoch AI2026-08-28
115GPT-3.5-turbo OpenAI0%Epoch AI2026-08-07
116GPT-4o mini OpenAI0%Epoch AI2026-07-15
117Granite 4.0 Micro IBM0%Epoch AI2026-08-30
118Llama 2-13B Meta0%Epoch AI2026-08-28
119Llama 2-7B Meta0%Epoch AI2026-08-30
120Llama 3.1-8B Meta0%Epoch AI2026-08-27
121Llama 3.2 1B Meta0%Epoch AI2026-08-30
122Llama 3-8B Meta0%Epoch AI2026-08-28
123Mistral 7B Mistral AI0%Epoch AI2026-08-30
124Mistral Small 3 Mistral AI0%Epoch AI2026-08-30
125Phi 3 Mini 4k Instruct Microsoft0%Epoch AI2026-08-30
126Qwen2.5 32B Instruct Alibaba (Qwen)0%Epoch AI2026-08-30
127Qwen2.5 7B Instruct Alibaba (Qwen)0%Epoch AI2026-08-30
128Qwen3-1.7B Alibaba (Qwen)0%noneEpoch AI2026-08-28
129Qwen3.5-2B Alibaba (Qwen)0%noneEpoch AI2026-08-28

Compare the leaders

Other reasoning benchmarks

Frequently asked questions

What does Chess Puzzles measure?

Chess tactics puzzles solved from a text position, without an engine.

Which model has the highest Chess Puzzles score?

As of October 2026, GPT-6 Astra has the highest published Chess Puzzles score on Noometry at 72%, out of 129 models with results.

What is the best open-weight model on Chess Puzzles?

DeepSeek V4 Pro has the highest Chess Puzzles accuracy among open-weight models at 47%, ranking 13 of 129 overall.