# Best agent frameworks and SDKs (slim)

> OpenAI Agents SDK (AA), Pydantic AI (A) and Microsoft Agent Framework (A) lead the 10 ranked agent frameworks and SDKs. Picks by need, strengths, weaknesses and prices from the Anchor benchmark.

- Full: https://www.anchorterminal.com/best/frameworks/index.md (~5,500 tokens) · this version ~1,280 tokens · JSON https://www.anchorterminal.com/best/frameworks/index.json · canonical https://www.anchorterminal.com/best/frameworks/
- Index: https://www.anchorterminal.com/llms.txt · API: https://www.anchorterminal.com/api/v1/index.json · Updated: 2026-10-10

All 10 ranked agent frameworks and SDKs on the Anchor benchmark, with a pick for each need and where each one falls short. Scores come from public evidence, re-checked as vendors change.

- Ranked: 10 · agent-ready (BB or better): 9 · accept x402: 0 · hosted endpoints: 0
- Full ranked table: https://www.anchorterminal.com/categories/frameworks.md
- Head-to-head comparisons: https://www.anchorterminal.com/compare/frameworks/index.md (45)
- Methodology: https://www.anchorterminal.com/benchmark/index.md

## The shortlist

| # | Tool | Grade | Score | Best for | Price | Where |
| --- | --- | --- | --- | --- | --- | --- |
| 1 | [OpenAI Agents SDK](https://www.anchorterminal.com/tools/openai-agents-sdk.md) | AA | 86.2 | Teams that want hand-offs, guardrails and approvals with little code, and are happy with OpenAI's tracing or will turn it off. | Free · OSS | Library |
| 2 | [Pydantic AI](https://www.anchorterminal.com/tools/pydantic-ai.md) | A | 83.7 | Python teams that want typed, validated outputs and tools, many model providers and durable runs on an engine they already use. | Free · OSS | Library |
| 3 | [Microsoft Agent Framework](https://www.anchorterminal.com/tools/microsoft-agent-framework.md) | A | 82.2 | Teams on Azure or .NET, and for teams moving off AutoGen or Semantic Kernel, that want MCP, approval, checkpointed workflows and OpenTelemetry in one SDK. | Free · OSS | Library |
| 4 | [Docker Agent](https://www.anchorterminal.com/tools/docker-agent.md) | BB | 76.5 | Teams already on Docker who want agents defined in a file, shared through an OCI registry and run the same way in a terminal, in CI or behind an HTTP, MCP or A2A server. | Free · OSS | Library |
| 5 | [AgentOS](https://www.anchorterminal.com/tools/agentos.md) | BB | 75.3 | TypeScript teams that want long-term memory, personas, voice and channel adapters and multi-agent teams in one package, and that will pin versions. | Free · OSS | Library |
| 6 | [Agent Development Kit (ADK)](https://www.anchorterminal.com/tools/google-adk.md) | BB | 74.7 | Teams on Google Cloud or working outside Python, and for anyone who wants approval, guardrail callbacks and evaluation in one framework. | Free · OSS | Library |
| 7 | [Agno](https://www.anchorterminal.com/tools/agno.md) | BB | 72.8 | Python teams that want agents, teams and workflows plus a self-hosted server with REST, MCP, approvals and scheduling in one package. | Free · OSS | Library |
| 8 | [Claude Agent SDK](https://www.anchorterminal.com/tools/claude-agent-sdk.md) | BB | 72.1 | Agents that work on files, code and shells and want Claude Code's tool set and permission model without building them. | Free | Library |
| 9 | [LangGraph](https://www.anchorterminal.com/tools/langgraph.md) | BB | 70.6 | Long-running, stateful agents that pause for people and must survive restarts. | Free · OSS | Library |
| 10 | [CrewAI](https://www.anchorterminal.com/tools/crewai.md) | B | 67 | Multi-agent role-and-task workflows and for teams that want Flows for ordered, stateful steps. | Free · OSS | Library |

## Picks by need

- Highest score overall: [OpenAI Agents SDK](https://www.anchorterminal.com/tools/openai-agents-sdk.md), AA, 86.2/100 on the benchmark. Also [Pydantic AI](https://www.anchorterminal.com/tools/pydantic-ai.md), A, 83.7/100.
- Reliability: [Docker Agent](https://www.anchorterminal.com/tools/docker-agent.md), 88/100 on reliability, against 85 for the overall leader.
- Schema & documentation: [Pydantic AI](https://www.anchorterminal.com/tools/pydantic-ai.md), 100/100 on schema & documentation, against 95 for the overall leader.
- Security & auth: [Pydantic AI](https://www.anchorterminal.com/tools/pydantic-ai.md), 95/100 on security & auth, against 80 for the overall leader.
- The review panel's favourite: [Pydantic AI](https://www.anchorterminal.com/tools/pydantic-ai.md), 4/5 from 8 panel reviews.

## How to choose

- Telemetry sent by default: Check what the framework sends by default and how to switch it off, since traces can carry prompts, tool arguments and customer data to a third party.
- Durable state and resumption: Check where run state is stored and whether a run resumes after a crash mid-tool-call, because an agent that restarts from scratch can repeat payments or emails.
- Human approval enforcement: Check whether approval gates run in code or only appear in prompts, since a model can ignore a prompt rule but a code check is harder to get past.
- Multi-agent hand-offs and tracing: Check how hand-offs pass context and whether each step appears in traces, since a lost hand-off is hard to debug without a record of what each agent saw.

Each listing's verdict, strengths and weaknesses: https://www.anchorterminal.com/best/frameworks/index.md