Skip to content

Benchmark

Arena ELO - Longer Query

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

Models measured
345
Scores recorded
401
Sources
1
IQ weight
1.34

Score distribution

1,323.7
Mean
1,357.1
Median
137.1
Std dev
1,102.6
10th percentile
1,475.3
90th percentile
951.6
Min
951.61,549
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 345 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 Next 80B A3B Instruct· LMArena data1,402.7
152Step 3.5 Flash· LMArena data1,402.1
153Claude Opus 420250514· LMArena data1,401.3
154Inkling 2 Small· LMArena data1,401
155DeepSeek V3.1· LMArena data1,399.7
156Qwen 3 235B A22B2507, thinking· LMArena data1,399.6
157GPT-5 Chat· LMArena data1,398
158NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,398
159Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,397.9
160nova-experimental-chat26-01-10· LMArena data1,397.1
161DeepSeek V3.1 Terminus· LMArena data1,395.7
162GPT-5.3 Chat· LMArena data1,395.5
163Gemini 3.1 Flash Lite· LMArena data1,394.8
164nova-experimental-chat11-10· LMArena data1,394.3
165MiMo V2 FlashThinking· LMArena data1,393.5
166Grok 4.3· LMArena data1,392.5
167Qwen3.5 Flash· LMArena data1,391.8
168Claude Haiku 5.5High· LMArena data1,391.5
169DeepSeek R10528· LMArena data1,391.1
170Qwen3 VL 235Bthinking· LMArena data1,390.3
171Grok 4.1 FastReasoning· LMArena data1,390.3
172LongCat Flash Chat· LMArena data1,390.3
173Qwen3.5 35B A3B· LMArena data1,390.1
174nova-experimental-chat12-10· LMArena data1,389.3
175Qwen 3 235B A22BNon-Thinking· LMArena data1,388.7
176GPT 5High· LMArena data1,387.3
177Hunyuan T120250711· LMArena data1,384.9
178GPT 4.12025-04-14· LMArena data1,384.6
179Muse Glimmer· LMArena data1,382.1
180Solar Pro 4· LMArena data1,380.2
181Claude Sonnet 420250514· LMArena data1,379.5
182Qwen3 Coder· LMArena data1,378.2
183Qwen3 30B A3B Instruct2507· LMArena data1,378.1
184o12024-12-17· LMArena data1,377.5
185Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,373.1
186Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,373
187Claude Sonnet 3.7Thinking 32K, 20250219· LMArena data1,372.8
188Grok 3 MiniHigh· LMArena data1,371.7
189OpenAI o32025-04-16· LMArena data1,371.2
190nova-experimental-chat10-20· LMArena data1,370
191MiniMax M2.5· LMArena data1,366
192GLM 4.5 Air· LMArena data1,365.8
193Hunyuan TurboS20250416· LMArena data1,365.6
194GPT 5.4 NanoHigh· LMArena data1,365.2
195Nemotron 3 Super 120B A12B· LMArena data1,361.9
196Mistral Medium 32505· LMArena data1,360
197Granite 4.2 30B· LMArena data1,358.2
198Qwen2.5-Max· LMArena data1,357.7
199Grok 3 Minibeta· LMArena data1,357.3
200GLM 4.6v· LMArena data1,357.1

Showing 151–200 of 401 results.

Data credits: LMArena, CC BY 4.0 (licence)