Skip to content

Benchmark

Arena ELO - Math

Math benchmarks measure how well models solve maths problems. Higher scores are better.

Models measured
350
Scores recorded
405
Sources
1
IQ weight
1.24

Score distribution

1,316
Mean
1,367.4
Median
152.6
Std dev
1,089
10th percentile
1,474.8
90th percentile
837.8
Min
837.81,547.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.24 for chat models, measured over 350 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101DeepSeek V4 ProHigh, 20260813· LMArena data1,432.1
102GLM 4.6· LMArena data1,432
103Qwen3.5 122B-A10B· LMArena data1,431.9
104nova-experimental-chat11-10· LMArena data1,431.6
105Mistral Medium 3.5· LMArena data1,431.5
106Qwen3 235B A22B Instruct 25072507· LMArena data1,431.2
107Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,431.1
108MiMo V2 Omni· LMArena data1,430.8
109Qwen3.5 27B· LMArena data1,429.3
110Kimi K2 Thinking· LMArena data1,428.6
111Gemini 3.1 Flash Lite· LMArena data1,428
112GLM 4.5· LMArena data1,427.5
113Qwen3 VL 235B· LMArena data1,426.9
114DeepSeek V4 Flash· LMArena data1,426.2
115OpenAI o32025-04-16· LMArena data1,426
116nova-experimental-chat10-20· LMArena data1,426
117DeepSeek-V4-Flash PreviewHigh· LMArena data1,424.3
118GLM 4.7· LMArena data1,422.7
119Grok 4.1Thinking· LMArena data1,422.6
120DeepSeek V3.2Thinking· LMArena data1,422.3
121Grok 40709· LMArena data1,422.3
122Claude Opus 4.120250805· LMArena data1,422
123MiniMax M2.7· LMArena data1,421.9
124MiniMax M3· LMArena data1,421.6
125GPT-5.5 Instant· LMArena data1,421.3
126DeepSeek V3.2 ExpThinking· LMArena data1,421.2
127Claude Sonnet 4.520250929· LMArena data1,421
128nova-experimental-chat12-10· LMArena data1,420.8
129DeepSeek V3.1· LMArena data1,420.6
130Grok 4 FastNon-Reasoning· LMArena data1,418.6
131GPT 5.2· LMArena data1,418.2
132GPT-5.4 miniHigh· LMArena data1,418.2
133DeepSeek V3.2 Exp· LMArena data1,418
134Solar Pro 4· LMArena data1,417.5
135Grok 4.6High· LMArena data1,416.7
136Grok 4.1· LMArena data1,416.7
137Gemini 2.5 Flash Preview09-2025· LMArena data1,416.1
138DeepSeek V3.1Thinking· LMArena data1,416.1
139Mistral Large 3· LMArena data1,415.5
140Hunyuan T120250711· LMArena data1,415.4
141Qwen 3 235B A22B2507, thinking· LMArena data1,413.7
142NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,412.6
143Grok 4.7xHigh· LMArena data1,412.4
144GPT-4.5 Preview2025-02-27· LMArena data1,412.1
145Muse Glimmer· LMArena data1,412
146ERNIE-5.0-Preview1203· LMArena data1,411.3
147Qwen3 VL 235Bthinking· LMArena data1,410.8
148Gemini 2.5 Flash· LMArena data1,410.7
149GPT 5.1· LMArena data1,410
150GPT-6 LunaMax· LMArena data1,410

Showing 101–150 of 405 results.

Data credits: LMArena, CC BY 4.0 (licence)