reasoning benchmark
PaperBench
PaperBench is a benchmark for evaluating AI agents on their ability to replicate research papers. It tests models on complex, multi-step workflows involving code implementation, experimentation, and reproducing scientific results from academic publications.
Updated Aug 11, 2026
Models3
Model coverage3
MetricScore
EvidenceB
PaperBench Ranking
Higher score ranks better on this benchmark.
| 01 | AC | 93.0% | 100.0% | 3 | C | |
| 02 | MA | 63.5% | 50.0% | 3 | C | |
| 03 | MI | 52.6% | 0.0% | 3 | C |
PaperBench Score Distribution
A closer view of the leading scores on this benchmark.
PaperBench Highlights
The leading models and scores on this benchmark.
What is PaperBench?
What PaperBench measures and how its scores work.
PaperBench is a benchmark for evaluating AI agents on their ability to replicate research papers. It tests models on complex, multi-step workflows involving code implementation, experimentation, and reproducing scientific results from academic publications.
Scores are shown in ratio. This benchmark is not independently verified and has an evidence level of B.
- Family
- PaperBench
- Modality
- text
- Primary category
- reasoning
- Score direction
- higher
- LLMBoard eligible
- Yes
- Evaluation key
- paperbench|llm-stats-current
Benchmark scores retain their original unit. Overall score eligibility is shown separately.
FAQ
Common questions about PaperBench.
Which model scores highest on PaperBench?
Qwen3.8 Max is currently ranked first with 93.0%.
What does PaperBench measure?
PaperBench is a benchmark for evaluating AI agents on their ability to replicate research papers. It tests models on complex, multi-step workflows involving code implementation, experimentation, and reproducing scientific results from academic publications.
Is a higher score better?
Yes. Higher values rank better for this benchmark.
How many models are compared?
3 model results are currently shown.
Does this benchmark affect the overall score?
Yes. This benchmark can contribute to the current LLMBoard capability score.