# Agent frameworks and SDKs (slim)

> 6 agent frameworks & SDKs ranked by the Anchor benchmark. Leader OpenAI Agents SDK (AA). Libraries that run the agent loop. Tool calling, MCP, multi-agent hand-offs, durable state, human approval and tracing. Compared on what they do by default, including the telemetry they send.

- Full: https://www.anchorterminal.com/categories/frameworks.md (~1,800 tokens) · this version ~380 tokens · JSON https://www.anchorterminal.com/categories/frameworks.json · canonical https://www.anchorterminal.com/categories/frameworks
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-04

- Tools: 6 · agent-ready (BB+): 5 · accept x402: 0 · JSON: https://www.anchorterminal.com/api/v1/tools.json · by capability: https://www.anchorterminal.com/api/v1/capabilities.json
- Capabilities: agent.framework, agent.multi-agent, agent.durable, agent.mcp-client

| # | Tool | Grade | Score | Confidence | x402 | Auth | Where | Slim |
| --- | --- | --- | --- | --- | --- | --- | --- | --- |
| 1 | OpenAI Agents SDK | AA | 86.5 | high | no | API key | library | https://www.anchorterminal.com/tools/openai-agents-sdk.min.md |
| 7 | Pydantic AI | A | 80 | medium | no | None | library | https://www.anchorterminal.com/tools/pydantic-ai.min.md |
| 45 | Agent Development Kit (ADK) | BB | 74.9 | medium | no | None | library | https://www.anchorterminal.com/tools/google-adk.min.md |
| 71 | Claude Agent SDK | BB | 72.4 | medium | no | API key | library | https://www.anchorterminal.com/tools/claude-agent-sdk.min.md |
| 95 | LangGraph | BB | 70.6 | medium | no | None | library | https://www.anchorterminal.com/tools/langgraph.min.md |
| 149 | CrewAI | B | 67 | medium | no | None | library | https://www.anchorterminal.com/tools/crewai.min.md |