Skip to content

Benchmark

Arena ELO - Industry Writing and Literature and Language

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

Models measured
365
Scores recorded
422
Sources
1
IQ weight
1.17

Score distribution

1,297.8
Mean
1,328.7
Median
142.8
Std dev
1,086
10th percentile
1,453.9
90th percentile
809.1
Min
809.11,540.8
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.17 for chat models, measured over 365 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 3.5· LMArena data1,382.8
152GPT-6 LunaMax· LMArena data1,382.6
153Claude Opus 420250514· LMArena data1,381.4
154Kimi K2 Thinking· LMArena data1,380.7
155Grok 4 FastReasoning· LMArena data1,380.3
156Qwen3.5 27B· LMArena data1,380
157nova-experimental-chat12-10· LMArena data1,379.8
158Inkling 2 Small· LMArena data1,379.8
159GPT-5.3 Chat· LMArena data1,378.5
160GPT 5High· LMArena data1,378.3
161MiniMax M2.7· LMArena data1,375.7
162NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,375.5
163Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,375
164MiniMax-M2.1 Preview· LMArena data1,372.9
165Step 3.5 Flash· LMArena data1,371.9
166OpenAI o32025-04-16· LMArena data1,371.4
167GPT 4.12025-04-14· LMArena data1,371.3
168nova-experimental-chat11-10· LMArena data1,370.4
169Qwen3.5 Flash· LMArena data1,369.2
170Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,368.6
171LongCat Flash Chat· LMArena data1,367
172Mistral Large 4· LMArena data1,366.9
173Qwen 3 235B A22BNon-Thinking· LMArena data1,366.2
174Claude Haiku 5.5High· LMArena data1,366
175Qwen3.5 35B A3B· LMArena data1,365.7
176DeepSeek V30324· LMArena data1,364.6
177DeepSeek R1· LMArena data1,364.4
178Qwen3 VL 235Bthinking· LMArena data1,364.3
179MiMo V2 FlashThinking· LMArena data1,362.7
180o12024-12-17· LMArena data1,360
181Inkling Small· LMArena data1,359.6
182Muse Glimmer· LMArena data1,358.6
183Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,356.9
184Qwen3 Next 80B A3B Instruct· LMArena data1,355
185GLM 4.5 Air· LMArena data1,353.8
186Mistral Medium 32505· LMArena data1,351.7
187Hunyuan TurboS20250416· LMArena data1,351.1
188Claude Sonnet 420250514· LMArena data1,350.8
189nova-experimental-chat10-20· LMArena data1,349.8
190Gemma 3 27B IT· LMArena data1,347.3
191Gemini 2.0 Flash001· LMArena data1,347
192Qwen2.5-Max· LMArena data1,346.1
193Grok 3 MiniHigh· LMArena data1,345.6
194nova-experimental-chat26-01-10· LMArena data1,345.6
195GLM 4.6v· LMArena data1,345.4
196MiniMax M2.5· LMArena data1,345.1
197Kimi K2 Preview0905· LMArena data1,345.1
198Grok 3 Minibeta· LMArena data1,345
199GPT-5 miniHigh· LMArena data1,344.5
200Solar Pro 4· LMArena data1,344.4

Showing 151–200 of 422 results.

Data credits: LMArena, CC BY 4.0 (licence)