Skip to content

Benchmark

Arena ELO - Exclude Ties

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.28

Score distribution

1,247.7
Mean
1,316.3
Median
217.2
Std dev
908.5
10th percentile
1,465.2
90th percentile
533.7
Min
533.71,565.6
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.28 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
151Qwen3.5 27B· LMArena data1,391.3
152Grok 4 FastNon-Reasoning· LMArena data1,390.2
153Gemini 2.5 Flash Preview09-2025· LMArena data1,388.9
154MiniMax M2.7· LMArena data1,387.5
155Hunyuan Vision 1.5 Thinking· LMArena data1,387.3
156GPT 5High· LMArena data1,387.3
157Inkling 2 Small· LMArena data1,387
158GPT-5 Chat· LMArena data1,383.9
159Step 3.5 Flash· LMArena data1,383.7
160Grok 4.7xHigh· LMArena data1,381.3
161Hunyuan T120250711· LMArena data1,380
162Grok 4 FastReasoning· LMArena data1,379.8
163Qwen3 VL 235Bthinking· LMArena data1,378.7
164GPT-6 SolMax· LMArena data1,378.1
165Claude Haiku 5.5High· LMArena data1,377.9
166Grok 4.3· LMArena data1,377.8
167Qwen3.5 Flash· LMArena data1,376.3
168Claude Haiku 4.520251001· LMArena data1,376
169Qwen3.5 35B A3B· LMArena data1,373.5
170nova-experimental-chat10-20· LMArena data1,372.7
171nova-experimental-chat26-01-10· LMArena data1,371.1
172Qwen 3 235B A22BNon-Thinking· LMArena data1,370.6
173MiMo V2 FlashThinking· LMArena data1,370.6
174GPT-6 LunaMax· LMArena data1,368.2
175GPT-5.3 Chat· LMArena data1,368
176MiniMax-M2.1 Preview· LMArena data1,367.7
177Muse Glimmer· LMArena data1,365.9
178Solar Pro 4· LMArena data1,361.9
179Qwen3 30B A3B Instruct2507· LMArena data1,357.7
180GLM 4.5 Air· LMArena data1,357.4
181GPT 4.12025-04-14· LMArena data1,355.9
182Kimi K2 Preview0905· LMArena data1,351.3
183Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,349.8
184Hunyuan TurboS20250416· LMArena data1,349.1
185GLM 4.6v· LMArena data1,347.5
186Claude Opus 4Thinking 16K, 20250514· LMArena data1,346.7
187DeepSeek R1· LMArena data1,345.8
188Nemotron 3 Super 120B A12B· LMArena data1,345.5
189DeepSeek V30324· LMArena data1,345.3
190GPT 5.4 NanoHigh· LMArena data1,345
191GPT-5 miniHigh· LMArena data1,342.8
192Mistral Medium 32505· LMArena data1,340.3
193Kimi K2 Preview0711· LMArena data1,338.2
194Qwen2.5-Max· LMArena data1,336.8
195Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,336.5
196Grok 3 MiniHigh· LMArena data1,335.1
197Qwen3 Next 80B A3Bthinking· LMArena data1,334.5
198o12024-12-17· LMArena data1,333.6
199Qwen 3 235B A22B· LMArena data1,332
200Claude Opus 420250514· LMArena data1,331.7

Showing 151–200 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)