Skip to content

Benchmark

Arena ELO - Polish

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

Models measured
207
Scores recorded
242
Sources
1
IQ weight
2.07

Score distribution

1,386.8
Mean
1,405.7
Median
84.1
Std dev
1,270
10th percentile
1,481.7
90th percentile
1,072
Min
1,0721,546.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 2.07 for chat models, measured over 207 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151LongCat-Flash-Chat Exp2602· LMArena data1,377.9
152Kimi K2.5Instant· LMArena data1,377.9
153Kimi K2 Preview0905· LMArena data1,377.6
154Qwen3.5 Flash· LMArena data1,376.7
155Qwen3.5 27B· LMArena data1,375.4
156Kimi K2 Preview0711· LMArena data1,371.8
157MiniMax-M2.1 Preview· LMArena data1,371.2
158Mistral Medium 32505· LMArena data1,370.2
159Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,370.1
160DeepSeek V30324· LMArena data1,368.7
161GLM 4.5 Air· LMArena data1,367.8
162GPT 5.4 NanoHigh· LMArena data1,365.4
163GPT 4o2024-05-13· LMArena data1,364.6
164Qwen3 VL 235Bthinking· LMArena data1,364.2
165Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,361.4
166Qwen2.5-Max· LMArena data1,361
167Qwen 3 235B A22B· LMArena data1,358.8
168o4 Mini2025-04-16· LMArena data1,357.5
169Grok 3 MiniHigh· LMArena data1,356
170Claude Opus 4Thinking 16K, 20250514· LMArena data1,355.6
171GPT OSS 120B· LMArena data1,353.2
172Qwen3 30B A3B Instruct2507· LMArena data1,353.1
173Grok 3 Minibeta· LMArena data1,351.9
174Claude 3 Opus20240229· LMArena data1,351.8
175INTELLECT-3· LMArena data1,348.9
176Gemini 2.0 Flash001· LMArena data1,348.7
177Nemotron 3 Super 120B A12B· LMArena data1,346.7
178Gemma 3 27B IT· LMArena data1,345.1
179Qwen3 Next 80B A3Bthinking· LMArena data1,344.8
180Qwen3 Coder· LMArena data1,344.6
181Claude Haiku 4.520251001· LMArena data1,343.7
182Claude Opus 420250514· LMArena data1,343.6
183MiniMax M2.5· LMArena data1,333.9
184MiniMax M1· LMArena data1,329.4
185Claude 3.5 Sonnet20240620· LMArena data1,328.4
186Nova 2 Lite· LMArena data1,327.2
187GPT-4 Turbo Preview1106· LMArena data1,326.4
188GPT-4.1 mini2025-04-14· LMArena data1,326
189Command A03-2025· LMArena data1,324.4
190Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,324.1
191Llama 3.3 Nemotron Super 49B v1.5· LMArena data1,323
192Claude Sonnet 420250514· LMArena data1,321.6
193Mistral Small 3.22506· LMArena data1,317
194GPT-5 nanoHigh· LMArena data1,313.3
195Ling Flash 2.0· LMArena data1,311.4
196MiniMax M2· LMArena data1,310.8
197Trinity Large Thinking· LMArena data1,308.6
198Step-1o Turbo202506· LMArena data1,303.8
199Nemotron 3 Nano 30B A3BBF16· LMArena data1,303.4
200GLM 4.5V· LMArena data1,302.3

Showing 151–200 of 242 results.

Data credits: LMArena, CC BY 4.0 (licence)