Skip to content

Benchmark

Arena ELO - Hard Prompts

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

Models measured
366
Scores recorded
423
Sources
1
IQ weight
1.24

Score distribution

1,308.5
Mean
1,356.7
Median
162.3
Std dev
1,065.4
10th percentile
1,480.3
90th percentile
727.9
Min
727.91,550.7
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.24 for chat models, measured over 366 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
101LongCat-Flash-Chat Exp2602· LMArena data1,439.2
102Gemma 4 26B A4B· LMArena data1,438.7
103Gemini 3 FlashMinimal· LMArena data1,437.9
104ERNIE-5.0-Preview1203· LMArena data1,437.8
105Mistral Medium 3.5· LMArena data1,436.4
106Grok 4.1· LMArena data1,435.7
107Mistral Large 4· LMArena data1,435.5
108GPT 5.1· LMArena data1,434.5
109DeepSeek V3.2· LMArena data1,433.8
110Grok 3 Preview02-24· LMArena data1,433.5
111Qwen3 235B A22B Instruct 25072507· LMArena data1,433.4
112Grok 4.1Thinking· LMArena data1,433.1
113LongCat Flash Chat· LMArena data1,432.5
114DeepSeek-V4-Flash PreviewHigh· LMArena data1,431.9
115DeepSeek V3.2Thinking· LMArena data1,430.9
116Mistral Large 3· LMArena data1,430.7
117DeepSeek V3.2 Exp· LMArena data1,429.4
118GLM 4.5· LMArena data1,429.3
119Qwen3 VL 235B· LMArena data1,428.2
120GPT 5.2High· LMArena data1,428.1
121Kimi K2 Thinking· LMArena data1,427.8
122Qwen3 Next 80B A3B Instruct· LMArena data1,427.6
123nova-experimental-chat12-10· LMArena data1,427.2
124DeepSeek V3.2 ExpThinking· LMArena data1,426.8
125GPT-5.5 Instant· LMArena data1,426.6
126GPT 5.2· LMArena data1,426.6
127DeepSeek V3.1 TerminusThinking· LMArena data1,426.1
128Mistral Medium 3.12508· LMArena data1,425.7
129Inkling Small· LMArena data1,424.9
130GPT-6.2 AstraMax· LMArena data1,424.8
131GPT-5.4 miniHigh· LMArena data1,424.7
132ChatGPT-4o Latest20250326· LMArena data1,424.2
133Qwen3 Max2025-09-23· LMArena data1,423.1
134Gemini 2.5 Flash· LMArena data1,422
135MiniMax M2.7· LMArena data1,421.8
136nova-experimental-chat11-10· LMArena data1,421.6
137Qwen3.5 122B-A10B· LMArena data1,421.5
138Inkling 2 Small· LMArena data1,420.8
139GPT-6 SolMax· LMArena data1,420.6
140MiMo V2 FlashNon-Thinking· LMArena data1,420.2
141ERNIE-5.0-Preview1022· LMArena data1,420.2
142Claude Haiku 4.520251001· LMArena data1,419.9
143GPT-6.2 SolMax· LMArena data1,419.2
144Hunyuan Vision 1.5 Thinking· LMArena data1,417.2
145DeepSeek V3.1Thinking· LMArena data1,416.6
146Qwen 3 235B A22B2507, thinking· LMArena data1,416
147GPT 5High· LMArena data1,415.7
148Claude Haiku 5.5High· LMArena data1,415.5
149NVIDIA Nemotron 3.5 Super VL 120BNVFP4· LMArena data1,415.5
150DeepSeek R10528· LMArena data1,415.1

Showing 101–150 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)