Benchmarks
Where frontier models fall short, measured.
- 80%
- 73%
- 67%
- 60%
- 60%
- 53%
- 33%
- 20%
- 47%
- 43%
- 43%
- 37%
- 37%
- 37%
- 37%
- 30%
- 27%
- 23%
- 3%
- November 8, 2026Real Unreal BenchLarge multi-system changes in complete, private Unreal Engine 5 games: fixing bugs, correcting work and building features across code, maps, materials and assets. A run is scored only on how the running game behaves, by deterministic checks; no model judges pass or fail.
- November 4, 2026SaaS Marathon BenchCan a model carry a long piece of business work across the software a company really runs on? Each task crosses several apps among HubSpot, Attio, Notion, Gmail, Google Calendar, Drive, Docs and Sheets, Granola, Mercury, Clay and lemlist. Models work through the GUI, by computer use on the screen, and through the CLI, over the apps' REST APIs and MCP servers together. Only what's in the apps at the end counts.