MGSM leaderboard
Top 10 by Accuracy · 75 models listed by the source
chatjudge.ai
MGSM: top 10 by Accuracy
Bars from zero. Leader: Claude Opus 4.5 (Thinking) at 95.20%. Full values at chatjudge.ai.
0%
20%
40%
60%
80%
100%
Claude Opus 4.5 (Thinking) (Anthropic), rank 1: 95.20%
A
Claude Opus 4.5 (Thinking)
95.20%
Claude Opus 4.5 (Nonthinking) (Anthropic), rank 2: 94.76%
A
Claude Opus 4.5 (Nonthinkin…
94.76%
Claude Opus 4.1 (Thinking) (Anthropic), rank 3: 94.44%
A
Claude Opus 4.1 (Thinking)
94.44%
Claude Sonnet 4.5 (Thinking) (Anthropic), rank 4: 94.33%
A
Claude Sonnet 4.5
94.33%
Claude Opus 4.1 (Nonthinking) (Anthropic), rank 5: 94.22%
A
Claude Opus 4.1 (Nonthinkin…
94.22%
GPT 5.2 (OpenAI), rank 6: 94.00%
O
GPT 5.2
94.00%
Gemini 3 Pro (11/25) (Google), rank 7: 93.93%
G
Gemini 3 Pro
93.93%
Claude Opus 4 (Nonthinking) (Anthropic), rank 8: 93.78%
A
Claude Opus 4
93.78%
o4 Mini (OpenAI), rank 9: 93.42%
O
o4 Mini
93.42%
Gemini 3 Flash (12/25) (Google), rank 10: 93.31%
G
Gemini 3 Flash
93.31%
MGSM: top 10 by Accuracy
Bars from zero. Leader: Claude Opus 4.5 (Thinking) at 95.20%. Full values at chatjudge.ai.
0%
50%
100%
Claude Opus 4.5 (Thinking) (Anthropic), rank 1: 95.20%
A
Claude Opus 4.5 …
95.20%
Claude Opus 4.5 (Nonthinking) (Anthropic), rank 2: 94.76%
A
Claude Opus 4.5 …
94.76%
Claude Opus 4.1 (Thinking) (Anthropic), rank 3: 94.44%
A
Claude Opus 4.1 …
94.44%
Claude Sonnet 4.5 (Thinking) (Anthropic), rank 4: 94.33%
A
Claude Sonnet 4.5
94.33%
Claude Opus 4.1 (Nonthinking) (Anthropic), rank 5: 94.22%
A
Claude Opus 4.1 …
94.22%
GPT 5.2 (OpenAI), rank 6: 94.00%
O
GPT 5.2
94.00%
Gemini 3 Pro (11/25) (Google), rank 7: 93.93%
G
Gemini 3 Pro
93.93%
Claude Opus 4 (Nonthinking) (Anthropic), rank 8: 93.78%
A
Claude Opus 4
93.78%
o4 Mini (OpenAI), rank 9: 93.42%
O
o4 Mini
93.42%
Gemini 3 Flash (12/25) (Google), rank 10: 93.31%
G
Gemini 3 Flash
93.31%
A
Anthropic
O
OpenAI
G
Google