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
151Grok 4.7xHigh· LMArena data1,415.1
152DeepSeek V3.1· LMArena data1,414.5
153Qwen3.5 27B· LMArena data1,414.1
154nova-experimental-chat26-01-10· LMArena data1,413.9
155Grok 4 FastNon-Reasoning· LMArena data1,412.2
156MiniMax-M2.1 Preview· LMArena data1,410.6
157Step 3.5 Flash· LMArena data1,410.5
158Gemini 2.5 Flash Preview09-2025· LMArena data1,409.1
159Grok 40709· LMArena data1,408.7
160GPT-6 LunaMax· LMArena data1,408.5
161Grok 4.1 FastReasoning· LMArena data1,407.8
162Gemini 3.1 Flash Lite· LMArena data1,407.7
163Qwen3 VL 235Bthinking· LMArena data1,406.8
164DeepSeek V3.1 Terminus· LMArena data1,406.2
165GPT-5 Chat· LMArena data1,404.5
166Grok 4 FastReasoning· LMArena data1,404.4
167MiMo V2 FlashThinking· LMArena data1,404.1
168Qwen3.5 Flash· LMArena data1,403.6
169GPT-4.5 Preview2025-02-27· LMArena data1,403.4
170OpenAI o32025-04-16· LMArena data1,401.8
171Qwen3.5 35B A3B· LMArena data1,399.9
172Solar Pro 4· LMArena data1,399.7
173Claude Opus 4Thinking 16K, 20250514· LMArena data1,399.3
174Hunyuan T120250711· LMArena data1,399
175GPT-5.3 Chat· LMArena data1,398.8
176Qwen3 30B A3B Instruct2507· LMArena data1,397.6
177Grok 4.3· LMArena data1,396.3
178Muse Glimmer· LMArena data1,395.9
179nova-experimental-chat10-20· LMArena data1,395.9
180Qwen 3 235B A22BNon-Thinking· LMArena data1,393.1
181Nemotron 3 Super 120B A12B· LMArena data1,385.9
182GPT 4.12025-04-14· LMArena data1,384.1
183Kimi K2 Preview0905· LMArena data1,383.2
184GPT-5 miniHigh· LMArena data1,380.2
185GPT 5.4 NanoHigh· LMArena data1,380.1
186GLM 4.5 Air· LMArena data1,378.7
187Gemini 2.5 Flash-Lite PreviewNon-Thinking, 09-2025· LMArena data1,376.8
188Claude Opus 420250514· LMArena data1,376.3
189Granite 4.2 30B· LMArena data1,376.1
190Grok 3 MiniHigh· LMArena data1,374.4
191Hunyuan TurboS20250416· LMArena data1,374
192Gemini 2.5 Flash-Lite PreviewThinking, 06-17· LMArena data1,373.6
193Claude Sonnet 4Thinking 32K, 20250514· LMArena data1,372.4
194MiniMax M2.5· LMArena data1,372.1
195Qwen3 Coder· LMArena data1,371.7
196o12024-12-17· LMArena data1,371.4
197Qwen3 Next 80B A3Bthinking· LMArena data1,371.4
198GLM 4.6v· LMArena data1,367.3
199Kimi K2 Preview0711· LMArena data1,365.7
200o3 MiniHigh· LMArena data1,365.4

Showing 151–200 of 423 results.

Data credits: LMArena, CC BY 4.0 (licence)