Skip to content

Benchmark

Arena Hard

People compare model answers and choose which they prefer. Scores are percentages.

Models measured
22
Scores recorded
22
Sources
1
IQ weight
1.10

Score distribution

66.4%
Mean
73.3%
Median
20.6%
Std dev
37.9%
10th percentile
91.7%
90th percentile
30.5%
Min
30.5%95.6%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 1.10 for chat models, measured over 22 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1Qwen 3 235B A22B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim95.6%
2Qwen3 32B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim93.8%
3Qwen3 14B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91.7%
4Qwen3 30B A3B· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim91%
5Llama 3.3 Nemotron Super 49B v1· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim88.3%
6Mistral Small 24B Instruct2501· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim87.6%
7Qwen2.5 72B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim81.2%
8Phi 4 Reasoning Plus· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim79%
9DeepSeek V2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim76.2%
10Phi 4· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim75.4%
11Phi-4-reasoning· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim73.3%
12Jamba 1.5 Large· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim65.4%
13Mistral Smalllatest· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim58.3%
14Ministral 3 14B Instruct2512· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim55.1%
15Mistral Large 3· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim55.1%
16Qwen2.5 7B Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim52%
17Ministral 3 8b Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim50.9%
18Jamba 1.5 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim46.1%
19Mistral Small 3.2 24B Instruct2506· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim43.1%
20phi-3.5-moe-instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim37.9%
21Phi 4 Mini· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim32.8%
22Ministral 3 3b Instruct· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim30.5%

Data credits: Data by LLM Stats