Capacidad
Evaluaciones de agentes y uso de herramientas
Evaluaciones de agentes y uso de herramientas. Serie de demostración del equipo de datos de The Latent.
Evaluaciones de agentes y uso de herramientas
- SWE-bench Verified
- OSWorld
- GAIA
Ver datos
| Date | SWE-bench Verified | OSWorld | GAIA |
|---|---|---|---|
| 2023-08-01 | 60.74 | 40.16 | 53.87 |
| 2023-09-01 | 59.44 | 39.32 | 52.76 |
| 2023-10-01 | 61.1 | 40.37 | 54.11 |
| 2023-11-01 | 67.42 | 44.57 | 59.76 |
| 2023-12-01 | 65.9 | 43.59 | 58.5 |
| 2024-01-01 | 66.31 | 43.85 | 58.81 |
| 2024-02-01 | 65.81 | 43.56 | 58.47 |
| 2024-03-01 | 60.64 | 40.12 | 53.84 |
| 2024-04-01 | 64.14 | 42.38 | 56.82 |
| 2024-05-01 | 66.58 | 44.03 | 59.06 |
| 2024-06-01 | 67.47 | 44.59 | 59.78 |
| 2024-07-01 | 72.91 | 48.21 | 64.65 |
| 2024-08-01 | 69.64 | 46.1 | 61.89 |
| 2024-09-01 | 66.82 | 44.19 | 59.28 |
| 2024-10-01 | 68.18 | 45.1 | 60.52 |
| 2024-11-01 | 65.13 | 43.07 | 57.78 |
| 2024-12-01 | 69.96 | 46.22 | 61.94 |
| 2025-01-01 | 74.81 | 49.47 | 66.36 |
| 2025-02-01 | 72.99 | 48.27 | 64.76 |
| 2025-03-01 | 75.62 | 50.01 | 67.08 |
| 2025-04-01 | 72.03 | 47.68 | 64.03 |
| 2025-05-01 | 68.07 | 45.01 | 60.37 |
| 2025-06-01 | 72.79 | 48.12 | 64.53 |
| 2025-07-01 | 72.86 | 48.18 | 64.63 |
| 2025-08-01 | 76.77 | 50.72 | 67.99 |
| 2025-09-01 | 81.57 | 53.95 | 72.39 |
| 2025-10-01 | 76.41 | 50.57 | 67.88 |
| 2025-11-01 | 76.32 | 50.47 | 67.7 |
| 2025-12-01 | 75.2 | 49.76 | 66.79 |
| 2026-01-01 | 72.41 | 47.87 | 64.2 |
| 2026-02-01 | 79.96 | 52.83 | 70.82 |
| 2026-03-01 | 82.04 | 54.26 | 72.8 |
| 2026-04-01 | 82.23 | 54.36 | 72.91 |
| 2026-05-01 | 85.29 | 56.42 | 75.71 |
| 2026-06-01 | 78.59 | 52.02 | 69.85 |
| 2026-07-01 | 77.6 | 51.3 | 68.8 |
Acceso API · Pro próximamente
Vista previa de demostración hasta julio de 2026 (percent).
Metodología
Serie sintética de demostración generada para previsualizar el diseño del producto de datos.