Skip to main content

Benchmark profile

PaperBench

A research-reproduction benchmark that asks agents to recreate the contributions of AI papers from the paper alone.

Data verified 21 confirmed releases in the last 30 daysSee provider release alerts

Benchmark score on PaperBench — August 3, 2026

BenchLM mirrors the published score view for PaperBench. Qwen3.8 Max leads the public snapshot at 93.0%. BenchLM does not use these results to rank models overall.

1 modelCodingCurrentDisplay onlyUpdated August 3, 2026

Benchmark score table (1 model)

Score
1
Qwen3.8 MaxAlibaba · Closed
93.0%

About PaperBench

Year

2026

Tasks

AI research-paper reproduction

Format

Long-horizon agent evaluation

Difficulty

Frontier autonomous research and engineering

Qwen evaluates Qwen3.8-Max in PaperBench's BasicAgent setting under Code-Dev mode, using Claude Opus 4.6 as judge and averaging three runs of up to 12 hours. We keep this provider-run score display-only because the agent setup and judge are part of the result.

BenchLM freshness & provenance

Version

PaperBench 2026

Refresh cadence

Quarterly

Staleness state

Current

Question availability

Public benchmark set

CurrentDisplay only

BenchLM uses freshness metadata to decide whether a benchmark should still be treated as a strong differentiator, a benchmark to watch, or a display-only reference. For the full scoring policy, see the BenchLM methodology page.

FAQ

What does PaperBench measure?

A research-reproduction benchmark that asks agents to recreate the contributions of AI papers from the paper alone.

Which model scores highest on PaperBench?

Qwen3.8 Max by Alibaba currently leads with a score of 93.0% on PaperBench.

How many models are evaluated on PaperBench?

1 AI models have been evaluated on PaperBench on BenchLM.

Last updated: August 3, 2026 · BenchLM version PaperBench 2026

Know when it’s worth switching models

The model to choose, the cheaper alternative, and the release we would wait on.

Read a sample issue

Join 2,000+ readers.

One email each week. Unsubscribe anytime.