Skip to content

Benchmark

Arena ELO - Industry Legal and Government

People compare model answers and choose which they prefer. Higher scores are better.

Models measured
338
Scores recorded
394
Sources
1
IQ weight
1.34

Score distribution

1,329.7
Mean
1,368.1
Median
139.4
Std dev
1,104.4
10th percentile
1,475.2
90th percentile
895.4
Min
895.41,543.1
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.34 for chat models, measured over 338 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151MiMo V2 FlashNon-Thinking· LMArena data1,409.4
152Grok 4 FastNon-Reasoning· LMArena data1,408.3
153Step 3.5 Flash· LMArena data1,407.8
154DeepSeek V3.1 Terminus· LMArena data1,407.5
155DeepSeek V3.2 Exp· LMArena data1,406.9
156Qwen3 VL 235Bthinking· LMArena data1,406.6
157Qwen3.5 27B· LMArena data1,405.2
158Kimi K2 Preview0905· LMArena data1,402.7
159Grok 4.1 FastReasoning· LMArena data1,402.4
160MiniMax M2.7· LMArena data1,401.8
161Claude Haiku 4.520251001· LMArena data1,400
162GPT-6 SolMax· LMArena data1,399.7
163GPT 4.12025-04-14· LMArena data1,399.5
164Qwen3 Max2025-09-23· LMArena data1,399.4
165Qwen3.5 Flash· LMArena data1,397.9
166Qwen3.5 35B A3B· LMArena data1,397.5
167MiMo V2 FlashThinking· LMArena data1,397.5
168GLM 4.6v· LMArena data1,394.9
169GPT-5.3 Chat· LMArena data1,394.1
170Qwen 3 235B A22BNon-Thinking· LMArena data1,394
171GPT-6 LunaMax· LMArena data1,392.6
172nova-experimental-chat10-20· LMArena data1,391.9
173Grok 3 MiniHigh· LMArena data1,391.3
174Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,391.3
175Nemotron 3 Super 120B A12B· LMArena data1,391.2
176Muse Glimmer· LMArena data1,389.8
177Claude Haiku 5.5High· LMArena data1,389.7
178Qwen3 30B A3B Instruct2507· LMArena data1,386.6
179Solar Pro 4· LMArena data1,386.4
180Grok 4.3· LMArena data1,385.8
181Mistral Medium 32505· LMArena data1,384
182GLM 4.5 Air· LMArena data1,382.3
183Gemma 3 12B IT· LMArena data1,380.8
184DeepSeek V30324· LMArena data1,379.4
185nova-experimental-chat26-01-10· LMArena data1,379.3
186Gemma 3 27B IT· LMArena data1,375.4
187GPT-5 miniHigh· LMArena data1,375.2
188Qwen2.5-Max· LMArena data1,374.8
189NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,374.4
190Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,373.3
191Claude Opus 420250514· LMArena data1,372.8
192Grok 3 Minibeta· LMArena data1,371.8
193Hunyuan TurboS20250416· LMArena data1,371.2
194Gemini 2.0 Flash001· LMArena data1,371
195DeepSeek R1· LMArena data1,371
196o12024-12-17· LMArena data1,368.2
197GPT 5.4 NanoHigh· LMArena data1,368.1
198Claude Opus 4Thinking 16K, 20250514· LMArena data1,366.5
199MiniMax M2.5· LMArena data1,366.3
200Qwen3 Next 80B A3Bthinking· LMArena data1,364

Showing 151–200 of 394 results.

Data credits: LMArena, CC BY 4.0 (licence)