Skip to content

Benchmark

Arena ELO - Chinese

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

Models measured
340
Scores recorded
396
Sources
1
IQ weight
1.23

Score distribution

1,334.6
Mean
1,384.4
Median
172.2
Std dev
1,067.9
10th percentile
1,512.4
90th percentile
719
Min
7191,628.9
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.23 for chat models, measured over 340 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
151Claude Haiku 5.5High· LMArena data1,434.5
152GPT-6.2 SolMax· LMArena data1,434.2
153GPT-5.3 Chat· LMArena data1,433.7
154Qwen3 30B A3B Instruct2507· LMArena data1,432.6
155Nemotron 3 Super 120B A12B· LMArena data1,431.6
156MiniMax-M2.1 Preview· LMArena data1,430.5
157Qwen3 Max2025-09-23· LMArena data1,429.9
158MiMo V2 FlashThinking· LMArena data1,428.9
159Solar Pro 4· LMArena data1,428.6
160Grok 40709· LMArena data1,427
161Claude Opus 4.120250805· LMArena data1,426.9
162GLM 4.5 Air· LMArena data1,426
163Hunyuan T120250711· LMArena data1,425.8
164Claude Opus 4.1Thinking 16K, 20250805· LMArena data1,424.9
165Qwen 3 235B A22BNon-Thinking· LMArena data1,424.5
166Granite 4.2 30B· LMArena data1,424.1
167GLM 4.6v· LMArena data1,423.6
168GPT-6 LunaMax· LMArena data1,422.7
169GPT 5High· LMArena data1,422.1
170GPT-5 Chat· LMArena data1,421.6
171Grok 4.3· LMArena data1,421.2
172GPT-4.5 Preview2025-02-27· LMArena data1,421.1
173Ling Flash 2.0· LMArena data1,420.3
174Claude Haiku 4.520251001· LMArena data1,418
175Hunyuan TurboS20250416· LMArena data1,416.1
176Kimi K2 Preview0905· LMArena data1,414.5
177Mercury 2· LMArena data1,413.4
178Muse Glimmer· LMArena data1,412.9
179GPT-6 SolMax· LMArena data1,408.2
180Qwen3 Next 80B A3Bthinking· LMArena data1,406.3
181nova-experimental-chat10-20· LMArena data1,406.1
182Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,404.7
183GLM 4.7 Flash· LMArena data1,404
184Kimi K2 Preview0711· LMArena data1,400.8
185DeepSeek R1· LMArena data1,400.2
186Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,400
187Ring Flash 2.0· LMArena data1,398.8
188Step3· LMArena data1,396.6
189MiniMax M2.5· LMArena data1,394.1
190o12024-12-17· LMArena data1,394
191GPT 5.4 NanoHigh· LMArena data1,391.1
192DeepSeek V30324· LMArena data1,390.5
193GLM 4 Plus0111· LMArena data1,389
194Grok 3 MiniHigh· LMArena data1,388
195Claude Opus 4Thinking 16K, 20250514· LMArena data1,385.9
196GPT-5 miniHigh· LMArena data1,385.8
197GPT OSS 120B· LMArena data1,384.7
198Nemotron 3 Nano 30B A3BBF16· LMArena data1,384.4
199GPT 4.12025-04-14· LMArena data1,382
200Claude Opus 420250514· LMArena data1,381.9

Showing 151–200 of 396 results.

Data credits: LMArena, CC BY 4.0 (licence)