# ExploitBench → StealthMark

Calibration · https://benchgap.net/calibration/exploitbench/current/to/stealthmark/current · data retrieved 2026-11-18

StealthMark is estimated from ExploitBench with a Hill curve fitted on 11 models measured on both: y = 0.4603 + (1.2000 − 0.4603)·x^3.94 / (1.23275^3.94 + x^3.94), R² = 0.81, cross-validated error 6.1 pp. It is used for 7 estimates.

| Estimated model | ExploitBench | StealthMark | Source |
|---|---|---|---|
| [GPT-6.2 Astra](https://benchgap.net/model/gpt-6-2-astra.md) | 100.0% | 68.6% | estimated ± 6.1 pp, medium confidence |
| [GPT-6.2 Sol](https://benchgap.net/model/gpt-6-2-sol.md) | 100.0% | 68.6% | estimated ± 6.1 pp, medium confidence |
| [GPT-6 Cyber](https://benchgap.net/model/gpt-6-cyber.md) | 100.0% | 68.6% | estimated ± 6.1 pp, medium confidence |
| [Claude Mythos 5.2](https://benchgap.net/model/claude-mythos-5-2.md) | 95.0% | 65.5% | estimated ± 6.1 pp, medium confidence |
| [Claude Mythos 5](https://benchgap.net/model/claude-mythos-5.md) | 78.0% | 56.5% | estimated ± 6.1 pp, medium confidence |
| [Claude Mythos Preview](https://benchgap.net/model/claude-mythos-preview.md) | 69.0% | 52.8% | estimated ± 6.1 pp, medium confidence |
| [SWE-2 Flash](https://benchgap.net/model/swe-2-flash.md) | 34.8% | 46.5% | estimated ± 6.1 pp, medium confidence |

## Fitted on

| Model | ExploitBench | StealthMark |
|---|---|---|
| [GLM-5.3](https://benchgap.net/model/glm-5-3.md) | 54.4% | 52.9% |
| [GPT-5.6 Luna](https://benchgap.net/model/gpt-5-6-luna.md) | 33.2% | 37.8% |
| [GPT-5.6 Sol](https://benchgap.net/model/gpt-5-6-sol.md) | 73.5% | 53.7% |
| [GPT-5.6 Terra](https://benchgap.net/model/gpt-5-6-terra.md) | 52.9% | 45.8% |
| [GPT-6.1 Sol](https://benchgap.net/model/gpt-6-1-sol.md) | 99.7% | 67.8% |
| [GPT-6.1 Terra](https://benchgap.net/model/gpt-6-1-terra.md) | 97.1% | 60.4% |
| [GPT-6 Astra](https://benchgap.net/model/gpt-6-astra.md) | 100.0% | 74.6% |
| [Kimi K3.1 Pro](https://benchgap.net/model/kimi-3-1-pro.md) | 74.7% | 56.7% |
| [Kimi K3](https://benchgap.net/model/kimi-k3.md) | 32.0% | 50.6% |
| [MiMo-V2.6-Flash](https://benchgap.net/model/mimo-v2-6-flash.md) | 25.3% | 49.3% |
| [MiMo-V2.6-Pro](https://benchgap.net/model/mimo-v2-6-pro.md) | 47.9% | 47.9% |
