Skip to content

Benchmark

Arena ELO Style-Controlled

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.10

Score distribution

1,341.3
Mean
1,363.7
Median
121
Std dev
1,152.6
10th percentile
1,474.5
90th percentile
953.4
Min
953.41,527
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.10 for chat models, measured over 366 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151MiniMax M2.7· LMArena data1,414.6
152GPT 4.12025-04-14· LMArena data1,414.6
153Claude Opus 420250514· LMArena data1,414.3
154Claude Haiku 4.520251001· LMArena data1,413.9
155Qwen3 VL 235B· LMArena data1,413.2
156Inkling 2 Small· LMArena data1,412
157nova-experimental-chat26-01-10· LMArena data1,411.7
158Grok 3 Preview02-24· LMArena data1,411.2
159GLM 4.5· LMArena data1,411
160Grok 40709· LMArena data1,410.9
161Gemini 2.5 Flash· LMArena data1,409.2
162Qwen3.5 27B· LMArena data1,408.4
163Mistral Medium 3.12508· LMArena data1,407.9
164Inkling Small· LMArena data1,405.5
165Grok 4 FastReasoning· LMArena data1,405.1
166Gemini 2.5 Flash Preview09-2025· LMArena data1,403
167o12024-12-17· LMArena data1,402.3
168Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,402
169Qwen 3 235B A22BNon-Thinking· LMArena data1,402
170GPT 5.4 NanoHigh· LMArena data1,401
171LongCat Flash Chat· LMArena data1,400.3
172Qwen 3 235B A22B2507, thinking· LMArena data1,400.1
173Qwen3 Next 80B A3B Instruct· LMArena data1,398.6
174DeepSeek R1· LMArena data1,398.1
175Qwen3.5 Flash· LMArena data1,395.9
176DeepSeek V30324· LMArena data1,395.1
177Qwen3 VL 235Bthinking· LMArena data1,394.6
178Hunyuan Vision 1.5 Thinking· LMArena data1,394.3
179Qwen3.5 35B A3B· LMArena data1,394.1
180Step 3.5 Flash· LMArena data1,392.8
181nova-experimental-chat12-10· LMArena data1,392.6
182MiMo V2 FlashNon-Thinking· LMArena data1,391.7
183Claude Sonnet 420250514· LMArena data1,390.9
184o4 Mini2025-04-16· LMArena data1,390.8
185MiniMax M2.5· LMArena data1,390.7
186GPT-5 miniHigh· LMArena data1,390.2
187O1 Preview· LMArena data1,388.7
188Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,388.4
189Solar Pro 4· LMArena data1,387.5
190Qwen3 Coder· LMArena data1,387.4
191Mistral Medium 32505· LMArena data1,387.2
192Hunyuan T120250711· LMArena data1,387.2
193MiMo V2 FlashThinking· LMArena data1,386.3
194MiniMax-M2.1 Preview· LMArena data1,384.3
195Hunyuan TurboS20250416· LMArena data1,383
196GPT-4.1 mini2025-04-14· LMArena data1,382.7
197Qwen3 30B A3B Instruct2507· LMArena data1,382.2
198Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,379.3
199Trinity Large Preview· LMArena data1,379
200GLM 4.6v· LMArena data1,378.3

Showing 151–200 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)