Benchmark category
Agents evaluations, model coverage and rankings. Each benchmark keeps its original scale and methodology.
Data as of 2026-08-11
Select a benchmark to inspect model results, evidence fields and scoring direction.
| BrowseComp | BrowseComp | text | Score | 58 | featured | B | Yes |
| Terminal-Bench 2.0 | Terminal-Bench 2.0 | text | Score | 49 | featured | C | Yes |
| SWE-Bench Pro | SWE-Bench Pro | text | Score | 45 | featured | B | Yes |
| MCP Atlas | MCP Atlas | text | Score | 31 | featured | B | Yes |
| Toolathlon | Toolathlon | text | Score | 31 | featured | B | Yes |
| Finance Agent v2 | Finance Agent v2 | text | Score | 26 | featured | B | Yes |
| Terminal-Bench | Terminal-Bench | text | Score | 25 | featured | B | Yes |
| OSWorld-Verified | OSWorld-Verified | multimodal | Score | 23 | featured | B | Yes |
| t2-bench | t2-bench | text | Score | 23 | featured | B | Yes |
| DeepSWE 1.1 | DeepSWE 1.1 | text | Score | 20 | featured | B | Yes |
| OSWorld | OSWorld | multimodal | Score | 20 | featured | B | Yes |
| BFCL-v3 | BFCL-v3 | text | Score | 19 | featured | B | Yes |
| Terminal-Bench 2.1 | Terminal-Bench 2.1 | text | Score | 19 | featured | B | Yes |
| FrontierCode 1.1 | FrontierCode 1.1 | text | Score | 15 | featured | B | Yes |
| FrontierSWE | FrontierSWE | text | Score | 15 | featured | B | Yes |
| NL2Repo | NL2Repo | text | Score | 14 | featured | B | Yes |
| BFCL-V4 | BFCL-V4 | text | Score | 13 | featured | B | Yes |
| Claw-Eval | Claw-Eval | text | Score | 13 | featured | B | Yes |
| Legal Agent Benchmark | Legal Agent Benchmark | text | Score | 13 | featured | B | Yes |
| CyberGym | CyberGym | text | Score | 11 | featured | B | Yes |
| DeepSWE | DeepSWE | text | Score | 10 | featured | B | Yes |
| VITA-Bench | VITA-Bench | text | Score | 10 | featured | B | Yes |
Leading results from high-coverage benchmarks with at least two models.
How this category is assembled on llmboard.ai.
This page groups benchmarks whose primary or display category matches agents. It does not average incompatible metrics into a new category score.
Open an individual benchmark to inspect score direction, evidence level, participant count and model results.
Benchmarks are grouped by their primary and display categories.