Skip to content

Benchmark

Arena ELO - German

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

Models measured
273
Scores recorded
319
Sources
1
IQ weight
1.52

Score distribution

1,328.6
Mean
1,372.1
Median
138.3
Std dev
1,110.8
10th percentile
1,469.8
90th percentile
901.9
Min
901.91,585.2
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.52 for chat models, measured over 273 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 4.12025-04-14· LMArena data1,380.8
152Qwen3 30B A3B Instruct2507· LMArena data1,379.4
153Mistral Medium 32505· LMArena data1,377.2
154Grok 4.7xHigh· LMArena data1,376.9
155GLM 4.5 Air· LMArena data1,376.9
156Kimi K2 Preview0711· LMArena data1,374.2
157DeepSeek V30324· LMArena data1,373.8
158Claude Haiku 4.520251001· LMArena data1,373.1
159Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,372.8
160Claude Opus 420250514· LMArena data1,372.1
161Step3· LMArena data1,370.7
162Gemma 3 12B IT· LMArena data1,370.3
163Qwen3.5 35B A3B· LMArena data1,367.8
164GPT 5.4 NanoHigh· LMArena data1,366.7
165GPT-5 miniHigh· LMArena data1,365.2
166LongCat-Flash-Chat Exp2602· LMArena data1,364.9
167Gemma 3 27B IT· LMArena data1,362.3
168MiniMax M2.5· LMArena data1,362.1
169Qwen3 Next 80B A3Bthinking· LMArena data1,356
170Hunyuan TurboS20250416· LMArena data1,355
171Trinity Large Thinking· LMArena data1,354.4
172MiniMax M2· LMArena data1,354
173Gemini 2.0 Flash001· LMArena data1,353.2
174GPT OSS 120B· LMArena data1,353.2
175Solar Pro 4· LMArena data1,352.4
176MiniMax M1· LMArena data1,350.2
177GPT-4.1 mini2025-04-14· LMArena data1,350.1
178Qwen2.5-Max· LMArena data1,349.6
179Grok 3 MiniHigh· LMArena data1,349.4
180Nemotron 3 Nano 30B A3BBF16· LMArena data1,348.4
181Grok 3 Minibeta· LMArena data1,347.8
182Nova 2 Lite· LMArena data1,343.1
183Command A03-2025· LMArena data1,342.3
184Nemotron 3 Super 120B A12B· LMArena data1,341.1
185Qwen3 32B· LMArena data1,341.1
186GLM 4 Plus0111· LMArena data1,340.9
187Mistral Small 3.22506· LMArena data1,340.9
188o12024-12-17· LMArena data1,336.8
189o4 Mini2025-04-16· LMArena data1,335.5
190GLM 4.7 Flash· LMArena data1,334
191Claude Sonnet 420250514· LMArena data1,330.6
192GPT-5 nanoHigh· LMArena data1,326.3
193Qwen3 Coder· LMArena data1,324.1
194DeepSeek V3· LMArena data1,323.5
195Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,317.5
196Trinity Large Preview· LMArena data1,317.2
197Step-1o Turbo202506· LMArena data1,315.4
198QwQ 32B· LMArena data1,312.2
199Gemma 3n E4B IT· LMArena data1,312.1
200O1 Preview· LMArena data1,309.3

Showing 151–200 of 319 results.

Data credits: LMArena, CC BY 4.0 (licence)