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
101GPT 5.1High· LMArena data1,433.2
102Qwen3.6 Plus· LMArena data1,431.6
103GPT-5.2 Chat20260210· LMArena data1,430.9
104DeepSeek V3.1· LMArena data1,430.6
105Qwen3 235B A22B Instruct 25072507· LMArena data1,429.3
106GPT-5.5 Instant· LMArena data1,429
107Qwen3 VL 235B· LMArena data1,427.9
108Gemini 3 FlashMinimal· LMArena data1,427.2
109Grok 4.1· LMArena data1,425.9
110DeepSeek V3.2· LMArena data1,425.1
111Qwen3.5 122B-A10B· LMArena data1,424.8
112ERNIE-5.0-Preview1022· LMArena data1,423.8
113Gemini 3.1 Flash Lite· LMArena data1,423.5
114Grok 4.6High· LMArena data1,422.1
115MiMo V2 FlashNon-Thinking· LMArena data1,422
116Gemini 2.5 Flash· LMArena data1,421.5
117Claude Haiku 4.520251001· LMArena data1,421.5
118MiniMax M3· LMArena data1,421.1
119Grok 40709· LMArena data1,419.9
120Qwen3 Max2025-09-23· LMArena data1,419.9
121GPT 5.1· LMArena data1,419.3
122Step 3.5 Flash· LMArena data1,418.6
123DeepSeek-V4-Flash PreviewHigh· LMArena data1,418
124GPT-6.1 TerraMax· LMArena data1,417.7
125Grok 3 Preview02-24· LMArena data1,417.2
126Nemotron 3 Super 120B A12B· LMArena data1,416.9
127DeepSeek V3.2Thinking· LMArena data1,416.4
128Grok 4 FastNon-Reasoning· LMArena data1,416.2
129Gemma 4 26B A4B· LMArena data1,415.6
130MiMo V2.5· LMArena data1,415.5
131Solar Pro 4· LMArena data1,414.1
132Grok 4.1 FastReasoning· LMArena data1,412.7
133LongCat-Flash-Chat Exp2602· LMArena data1,412.7
134GPT 5.2High· LMArena data1,412.2
135Mistral Medium 3.5· LMArena data1,411.5
136DeepSeek R10528· LMArena data1,411.3
137DeepSeek V3.1Thinking· LMArena data1,411
138Qwen3.5 27B· LMArena data1,408.9
139DeepSeek V3.2 ExpThinking· LMArena data1,408.6
140MiniMax M2.7· LMArena data1,407.4
141Kimi K2 Thinking· LMArena data1,406.9
142GPT-5.4 miniHigh· LMArena data1,405.7
143Ling Flash 2.0· LMArena data1,405
144Gemini 2.5 Flash Preview09-2025· LMArena data1,404.2
145Grok 4 FastReasoning· LMArena data1,403.9
146Qwen3 30B A3B Instruct2507· LMArena data1,403.4
147Qwen 3 235B A22BNon-Thinking· LMArena data1,402.8
148Qwen3 VL 235Bthinking· LMArena data1,402.6
149Qwen3.5 Flash· LMArena data1,400.8
150GPT 5.2· LMArena data1,400.2

Showing 101–150 of 305 results.

Data credits: LMArena, CC BY 4.0 (licence)