# FORTRESS (adversarial, Public) leaderboard

external benchmark

> Every figure below is reproduced as its upstream source published it: nothing is modelled, estimated, interpolated or converted. `Unavailable` means no source published the value — it is never a zero. Each value carries its evidence state and the date it was observed.

Page: https://modelchoice.dev/leaderboards/benchmarks/external-fortress-adversarial-public

FORTRESS Public adversarial safety split. Every model the catalog carries a published FORTRESS (adversarial, Public) value for, ranked by that value.

## Benchmark record

| Field | Value |
| --- | --- |
| Category | external |
| Measure | Safety score (%) |
| Tasks | Public adversarial safety prompts from FORTRESS |
| Difficulty | Adversarial safety and refusal |
| Published by | [Introducing Inkling 2 Mini](https://thinkingmachines.ai/news/inkling-2-mini/) |

The public adversarial split of Scale AI FORTRESS, measuring safe refusal and mitigation on high-risk adversarial safety prompts.

## FORTRESS (adversarial, Public) ranking

1 model with a published FORTRESS (adversarial, Public) value, ordered by that value, highest first. Models the source has not scored on this benchmark are not listed — they are unmeasured, not last.

| Rank | Model | Provider | Safety score (%) | Evidence |
| --- | --- | --- | --- | --- |
| 1 | [Inkling 2 Mini](https://modelchoice.dev/models/inkling-2-mini) | Thinking Machines Lab | 89.5 | Observed 2026-11-21 · source benchlm:models |

Rows are ordered by the value the source published, highest first. The source does not state whether a higher value is the better result, so this document does not either: for a benchmark that measures a rate of failure, read the table from the bottom.
