| Benchmark | Model | Rank | Results |
|---|---|---|---|
| deep-research-agents-on-browsecomp | Atria Dawn Preview | #1 | Accuracy: 92.5 |
| agents-on-automationbench-public | Atria Dawn Preview | #3 | Success Rate: 53.8 |
| vulnerability-reproduction-on-cybergym | Atria Dawn Preview (Claude Code v2.1.119) | #3 | Success Rate: 86.5 |
| code-generation-on-swe-bench-pro | Atria Dawn Preview | #14 | Resolved: 59.6 |
| code-generation-on-terminal-bench-2-1 | Atria Dawn Preview (Claude Code v2.1.224) | #17 | Accuracy: 78.3 |