Skip to content

Benchmark

Arena ELO - Non English

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.20

Score distribution

1,288.7
Mean
1,334.4
Median
160.8
Std dev
1,032.5
10th percentile
1,449.6
90th percentile
781.3
Min
781.31,534.5
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.20 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
151GPT-5 Chat· LMArena data1,394.7
152Hunyuan T120250711· LMArena data1,394.4
153MiMo V2 FlashNon-Thinking· LMArena data1,391.8
154Grok 4.1 FastReasoning· LMArena data1,391.6
155Inkling 2 Small· LMArena data1,390.2
156Qwen3.5 27B· LMArena data1,389.4
157Grok 4 FastReasoning· LMArena data1,388.3
158Grok 4.7xHigh· LMArena data1,388
159GPT-6 SolMax· LMArena data1,387.3
160NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,387.1
161Muse Glimmer· LMArena data1,385.5
162Grok 4.3· LMArena data1,385.5
163Step 3.5 Flash· LMArena data1,385.3
164Claude Haiku 5.5High· LMArena data1,385.2
165Qwen3.5 Flash· LMArena data1,384.9
166Qwen 3 235B A22BNon-Thinking· LMArena data1,384.1
167Hunyuan Vision 1.5 Thinking· LMArena data1,383.1
168MiniMax M2.7· LMArena data1,382.5
169GPT-5.3 Chat· LMArena data1,382.3
170GPT-6 LunaMax· LMArena data1,381.2
171nova-experimental-chat26-01-10· LMArena data1,380.2
172Qwen3.5 35B A3B· LMArena data1,378.2
173MiniMax-M2.1 Preview· LMArena data1,377.6
174Qwen3 VL 235Bthinking· LMArena data1,376.9
175Claude Haiku 4.520251001· LMArena data1,376.3
176MiMo V2 FlashThinking· LMArena data1,373.5
177Kimi K2 Preview0905· LMArena data1,372.3
178nova-experimental-chat10-20· LMArena data1,371.9
179Qwen3 30B A3B Instruct2507· LMArena data1,371.6
180GPT 4.12025-04-14· LMArena data1,369.7
181Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,369.1
182GLM 4.5 Air· LMArena data1,366
183Hunyuan TurboS20250416· LMArena data1,362.8
184GPT-5 miniHigh· LMArena data1,362.7
185Solar Pro 4· LMArena data1,362
186Claude Opus 4Thinking 16K, 20250514· LMArena data1,361.6
187Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,360
188GPT 5.4 NanoHigh· LMArena data1,359.1
189GLM 4.6v· LMArena data1,359.1
190o12024-12-17· LMArena data1,358.5
191Kimi K2 Preview0711· LMArena data1,358.2
192DeepSeek V30324· LMArena data1,357.9
193DeepSeek R1· LMArena data1,356.5
194Nemotron 3 Super 120B A12B· LMArena data1,355.6
195Mistral Medium 32505· LMArena data1,353
196Grok 3 MiniHigh· LMArena data1,352.1
197Qwen2.5-Max· LMArena data1,352.1
198Claude Opus 420250514· LMArena data1,351.9
199GPT OSS 120B· LMArena data1,350.3
200Grok 3 Minibeta· LMArena data1,348.5

Showing 151–200 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)