# StrongREJECT (no jailbreak) (StrongREJECT)

> StrongREJECT scores how well a model refuses forbidden requests; this setting sends the harmful prompts without any jailbreak wrapper. Higher scores mean safer responses.

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

- Category: [external](/external)
- Last updated: November 11, 2026

## About StrongREJECT

- Year: 2024
- Tasks: Forbidden requests across harm categories
- Format: Safe-response score (%)
- Difficulty: Harmful-request refusal
- Paper: [A StrongREJECT for Empty Jailbreaks](https://arxiv.org/abs/2402.10260)

Thinking Machines reports StrongREJECT with no jailbreak in the Inkling 2 Small launch (November 5, 2026): DeepSeek V4.1 Flash 98.7%, Mistral Large 4 98.6%, Inkling 2 Small 97.8%, Muse Spark Open 97.6%, GLM 5.3 97.2% and GLM-5.3-Flash 96.8%.

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

## Leaderboard (1 models)

| Rank | Model | Creator | Score |
|------|-------|---------|-------|
| 1 | [Inkling 2 Small](/models/inkling-2-small) | Thinking Machines Lab | 97.8% |

## FAQ

### What does StrongREJECT measure?

StrongREJECT scores how well a model refuses forbidden requests; this setting sends the harmful prompts without any jailbreak wrapper. Higher scores mean safer responses.

### Which model scores highest on StrongREJECT?

Inkling 2 Small by Thinking Machines Lab currently leads with a score of 97.8% on StrongREJECT.

### How many models are evaluated on StrongREJECT?

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

### Does StrongREJECT affect BenchLM's overall score?

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