AI benchmarks and leaderboards are different things
Learn the difference between a benchmark, an evaluation run and a leaderboard, and use each to answer the right question.
Model reports
Arena-Hard
LLM-judge dependent, with known style and length bias.
Evaluation context and evidence
- Source
- Model reports
- Publisher
- Not recorded
- Benchmark release
- 2024-04-19
- Collected
- Not recorded
- Modality
- Not recorded
These are source-reported observations, not independent reruns by this site. Check the task version, evaluation protocol, unit and reporting date before comparing models. Results from different setups are not a universal ranking.
Original sources
Reported results 3
Shown in source observation order, not ranked. Dates retain the precision provided by the source.
| Model | Value | Unit | Protocol | Reported date and basis | Source |
|---|---|---|---|---|---|
| DeepSeek-R1 | 92.3 | percent | GPT-4-1106 judge | 2025-01-22 · document_publication | DeepSeek-R1 |
| DeepSeek-V3 | 85.5 | percent | GPT-4-1106 judge | 2024-12-27 · document_publication | DeepSeek-V3 |
| Qwen3-235B-A22B (Thinking) | 95.6 | percent | thinking mode | 2025-05-14 · document_publication | Qwen3 |