# DiligenceBench (Thoughtful), best harness (Diligence Bench)

> A financial due-diligence agent benchmark from Thoughtful, scoring how well a model researches and analyzes a company the way an investment analyst would.

Canonical page: https://benchlm.ai/benchmarks/diligencebench

- Category: [Agentic](/agentic)
- Last updated: November 13, 2026

## About Diligence Bench

- Year: 2026
- Tasks: Financial due-diligence research tasks
- Format: Score (%)
- Difficulty: Agentic financial analysis
- Paper: [Introducing Inkling 2 Small](https://thinkingmachines.ai/news/inkling-2-small/)

Thinking Machines attributes DiligenceBench to Thoughtful and reports best-harness scores in the Inkling 2 Small launch (November 5, 2026): Muse Spark Open 59.0%, Inkling 2 Small 58.1% and DeepSeek-V4.1-Flash-1017 50.4%.

Diligence Bench is currently displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.

## Leaderboard (1 models)

The one result on this page was reported by the lab that built the model.

| Rank | Model | Creator | Score | Who ran it |
|------|-------|---------|-------|-----------|
| 1 | [Inkling 2 Small](/models/inkling-2-small) | Thinking Machines Lab | 58.1% | Reported by Thinking Machines Lab |

## FAQ

### What does Diligence Bench measure?

A financial due-diligence agent benchmark from Thoughtful, scoring how well a model researches and analyzes a company the way an investment analyst would.

### Which model scores highest on Diligence Bench?

Inkling 2 Small by Thinking Machines Lab currently leads with a score of 58.1% on Diligence Bench.

### How many models are evaluated on Diligence Bench?

1 AI models have published results on Diligence Bench in the BenchLM catalog.

### Does Diligence Bench affect BenchLM's overall score?

Not directly. Diligence Bench is still displayed on BenchLM for reference, but it is excluded from the weighted scoring formula.