# CVE Rediscovery — AI model rankings

Anthropic internal evaluation: share of 1,829 published CVEs a model finds on its own from a snapshot of the project before the fix

2 tracked models have a published CVE Rediscovery score. Higher is better. Scores come from published lab reports and benchmark sources; recorded sources appear beside the scores.

## Ranking

| Rank | Model | Developer | Score | Source | Released |
| --- | --- | --- | --- | --- | --- |
| 1 | Claude Mythos 5.2 | Anthropic | 89.6% | Lab | Nov 9 2026 |
| 2 | Claude Mythos 5.1 | Anthropic | 78.3% | Lab | Sep 1 2026 |


---

Canonical page: https://aireleasetracker.com/benchmark/cve-rediscovery
Site index: https://aireleasetracker.com/llms.txt
Source: AI Release Tracker (https://aireleasetracker.com). Most benchmark scores come from lab launch material; gathered results identify the leaderboard that published them.