Skip to content

Benchmark

Arena ELO - Industry Entertainment and Sports and Media

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

Models measured
364
Scores recorded
421
Sources
1
IQ weight
1.17

Score distribution

1,279.8
Mean
1,314.7
Median
149.5
Std dev
1,058.5
10th percentile
1,441.1
90th percentile
784.8
Min
784.81,526.7
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 364 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151GPT-6 SolMax· LMArena data1,368.7
152Qwen3 VL 235B· LMArena data1,367.1
153Grok 4 FastReasoning· LMArena data1,366.5
154Qwen3.5 122B-A10B· LMArena data1,366.4
155GPT 4.12025-04-14· LMArena data1,364.9
156nova-experimental-chat11-10· LMArena data1,364.6
157GPT-5 Chat· LMArena data1,362.5
158NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,362.3
159MiniMax-M2.1 Preview· LMArena data1,362
160Qwen3.5 27B· LMArena data1,361.7
161LongCat Flash Chat· LMArena data1,360.4
162MiniMax M2.7· LMArena data1,359.6
163GPT-6 LunaMax· LMArena data1,359.4
164Step 3.5 Flash· LMArena data1,359.3
165o12024-12-17· LMArena data1,358.1
166Mistral Large 4· LMArena data1,357.6
167GPT-5.3 Chat· LMArena data1,357.5
168DeepSeek V30324· LMArena data1,357.2
169MiMo V2 FlashThinking· LMArena data1,355.3
170Inkling 2 Small· LMArena data1,355
171Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,354.8
172Hunyuan T120250711· LMArena data1,353.4
173Inkling Small· LMArena data1,353.1
174Claude Opus 420250514· LMArena data1,350.3
175Claude Haiku 5.5High· LMArena data1,349.7
176Qwen 3 235B A22BNon-Thinking· LMArena data1,349.2
177Qwen3 Next 80B A3B Instruct· LMArena data1,349
178Qwen3.5 35B A3B· LMArena data1,348
179Qwen3.5 Flash· LMArena data1,347.7
180DeepSeek R1· LMArena data1,346.7
181nova-experimental-chat26-01-10· LMArena data1,346.6
182nova-experimental-chat12-10· LMArena data1,345.8
183Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,344.6
184Muse Glimmer· LMArena data1,343.8
185GLM 4.6v· LMArena data1,342.9
186nova-experimental-chat10-20· LMArena data1,342.8
187Kimi K2 Preview0905· LMArena data1,341.8
188GPT-5 miniHigh· LMArena data1,340.3
189GLM 4.5 Air· LMArena data1,339.3
190Grok 3 MiniHigh· LMArena data1,335.7
191Nemotron 3 Super 120B A12B· LMArena data1,335.2
192Mistral Medium 32505· LMArena data1,335
193Qwen3 30B A3B Instruct2507· LMArena data1,334.5
194Hunyuan TurboS20250416· LMArena data1,333.4
195Gemini 2.0 Flash001· LMArena data1,332.9
196Kimi K2 Preview0711· LMArena data1,332.7
197Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,331.9
198Grok 3 Minibeta· LMArena data1,331.9
199Step3· LMArena data1,331.5
200O1 Preview· LMArena data1,331

Showing 151–200 of 421 results.

Data credits: LMArena, CC BY 4.0 (licence)