Back to directory
Testing & Review
AgentBench: Evaluating LLMs as Agents
Hugging Face paper page on a benchmark to evaluate LLMs agents
Compatibility
Agent support matrix
3 supported
| Agent | Status |
|---|---|
| Claude Code | Supported |
| OpenCode | Not listed |
| Cursor | Supported |
| MCP | Not listed |
| GitHub Copilot | Not listed |
| Windsurf | Not listed |
| Codex | Supported |
| Gemini CLI |