Skip to content

Benchmark

Arena ELO - Spanish

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

Models measured
259
Scores recorded
305
Sources
1
IQ weight
1.90

Score distribution

1,355.9
Mean
1,399.3
Median
117.6
Std dev
1,167.6
10th percentile
1,470
90th percentile
1,007
Min
1,0071,522.3
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.90 for chat models, measured over 259 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 data1,399.3
152Grok 4.3· LMArena data1,399.3
153Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,396
154Claude Haiku 5.5High· LMArena data1,395.8
155GPT-6 LunaMax· LMArena data1,395.7
156MiniMax-M2.1 Preview· LMArena data1,395.6
157Grok 4.7xHigh· LMArena data1,395
158OpenAI o32025-04-16· LMArena data1,394.8
159GPT-6 SolMax· LMArena data1,394.4
160Qwen3.5 35B A3B· LMArena data1,393.7
161MiMo V2 FlashThinking· LMArena data1,393.5
162nova-experimental-chat26-01-10· LMArena data1,390.1
163Claude Opus 4Thinking 16K, 20250514· LMArena data1,389.2
164GPT OSS 120B· LMArena data1,389
165Qwen 3 235B A22B2507, thinking· LMArena data1,388.6
166GPT 5High· LMArena data1,387.9
167Kimi K2 Preview0905· LMArena data1,386.3
168GLM 4.5 Air· LMArena data1,385.5
169Ring Flash 2.0· LMArena data1,384.9
170Grok 3 Minibeta· LMArena data1,380.8
171nova-experimental-chat10-20· LMArena data1,380
172DeepSeek R1· LMArena data1,378.6
173Nemotron 3 Nano 30B A3BBF16· LMArena data1,376.9
174Qwen2.5-Max· LMArena data1,376.8
175GPT 4.12025-04-14· LMArena data1,376.1
176Nova 2 Lite· LMArena data1,375.5
177Step3· LMArena data1,371.4
178GPT 5.4 NanoHigh· LMArena data1,371.3
179GPT-5.3 Chat· LMArena data1,370.3
180Qwen 3 235B A22B· LMArena data1,368.5
181Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,364.2
182Gemini 2.0 Flash001· LMArena data1,363.7
183Mistral Medium 32505· LMArena data1,363.2
184GPT-5 nanoHigh· LMArena data1,361.7
185Grok 3 MiniHigh· LMArena data1,360.3
186Qwen3 Coder· LMArena data1,358.2
187Trinity Large Preview· LMArena data1,357.7
188Qwen3 Next 80B A3Bthinking· LMArena data1,357.5
189DeepSeek V30324· LMArena data1,357.5
190Claude Opus 420250514· LMArena data1,356.7
191Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,356.6
192GPT-5 miniHigh· LMArena data1,354.9
193Nemotron 3.5 Lightning 30B A3BNVFP4· LMArena data1,353.7
194MiniMax M1· LMArena data1,353.6
195MiniMax M2.5· LMArena data1,353.1
196DeepSeek V3· LMArena data1,352.2
197QwQ 32B· LMArena data1,351.4
198GLM 4.7 Flash· LMArena data1,350.8
199Claude Sonnet 420250514· LMArena data1,349.4
200Gemma 3 27B IT· LMArena data1,348.8

Showing 151–200 of 305 results.

Data credits: LMArena, CC BY 4.0 (licence)