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
151Mistral Medium 3.12508· LMArena data1,409
152GPT-5 Chat· LMArena data1,407.9
153Qwen3.5 Flash· LMArena data1,407.5
154Grok 4.1 FastReasoning· LMArena data1,407.3
155ChatGPT-4o Latest20250326· LMArena data1,407
156Step 3.5 Flash· LMArena data1,406.4
157GPT 5.4 NanoHigh· LMArena data1,405.3
158ERNIE-5.0-Preview1022· LMArena data1,405.3
159Grok 4 FastReasoning· LMArena data1,403.6
160Qwen3.5 35B A3B· LMArena data1,403.5
161DeepSeek V3.1 Terminus· LMArena data1,401.9
162nova-experimental-chat26-01-10· LMArena data1,401.5
163DeepSeek R10528· LMArena data1,400.1
164GPT-6 SolMax· LMArena data1,399.7
165Qwen3 32B· LMArena data1,398.5
166Qwen3 Next 80B A3Bthinking· LMArena data1,398.4
167GPT 5High· LMArena data1,398
168Claude Haiku 4.520251001· LMArena data1,397.7
169MiniMax-M2.1 Preview· LMArena data1,396.9
170Qwen 3 235B A22BNon-Thinking· LMArena data1,396.8
171Kimi K2 Preview0905· LMArena data1,396.8
172o3 MiniHigh· LMArena data1,395.8
173GLM 4.5 Air· LMArena data1,395.4
174MiMo V2 FlashNon-Thinking· LMArena data1,395.3
175Qwen3 30B A3B Instruct2507· LMArena data1,395.3
176Qwen 3 235B A22B· LMArena data1,394.5
177Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,394.3
178DeepSeek R1· LMArena data1,392.5
179Grok 3 Preview02-24· LMArena data1,390.6
180Claude Opus 4Thinking 16K, 20250514· LMArena data1,390.5
181GPT-5.3 Chat· LMArena data1,389
182o4 Mini2025-04-16· LMArena data1,388.8
183GPT OSS 120B· LMArena data1,388.7
184o12024-12-17· LMArena data1,388.4
185Grok 4.3· LMArena data1,387.1
186Grok 3 MiniHigh· LMArena data1,385.8
187Nemotron 3 Super 120B A12B· LMArena data1,379.4
188MiMo V2 FlashThinking· LMArena data1,378.5
189GPT-5 miniHigh· LMArena data1,378.4
190MiniMax M2.5· LMArena data1,376.8
191INTELLECT-3· LMArena data1,375.5
192Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,374.8
193o3 Mini· LMArena data1,374.2
194Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,373.2
195Claude Opus 420250514· LMArena data1,373.1
196O1 Preview· LMArena data1,373
197Grok 3 Minibeta· LMArena data1,372.4
198DeepSeek V30324· LMArena data1,372.1
199Nemotron 3 Nano 30B A3BBF16· LMArena data1,370.9
200GPT 4.12025-04-14· LMArena data1,369.6

Showing 151–200 of 405 results.

Data credits: LMArena, CC BY 4.0 (licence)