Thinking Machines Inkling-Small Launch - FORTRESS Benign: leaderboard
FORTRESS benign look-alike prompts (Scale AI) run by Thinking Machines: the rate at which a model still answers safe requests that resemble harmful ones, so higher means less over-refusal. Thinking Machines' Inkling-Small launch post (July 2026), Safety rows of the full comparison table; percent, higher is better; Inkling models at effort 0.99.
Metric: Benign-Request Answer Rate (%). Source: thinkingmachines.ai. Saturation forecast: Estimated already saturated. 10 models tracked.
Top models
| # | Model | Score | Overall rank |
|---|---|---|---|
| 1 | DeepSeek V4 Flash | 99.2 | #155 |
| 2 | GPT-5.6 Luna | 97.8 | #79 |
| 3 | Gemini 3.5 Flash Lite | 95.5 | #174 |
| 4 | Qwen 3.5 397B A17B | 95.4 | #154 |
| 5 | MiMo-V2.5 | 94.6 | #196 |
| 6 | Claude Haiku 4.5 | 94.1 | #287 |
| 7 | Nemotron 3 Ultra | 90.6 | #245 |
| 8 | MiniMax-M2.7 | 90.1 | #270 |
No result here: #3 Claude Opus 5.5, #6 GPT-6 Astra, #7 GPT-6.1 Sol.
Interactive version: theaggregate.ai/benchmark?slug=thinking-machines-inkling-small-launch-fortress-benign · How It Works · Data refreshed daily, snapshot 2026-11-05.