LangChain / LangGraph + LangSmith
Building custom, stateful agents in code and tracing, evaluating and deploying them.
Anthropic's pre-built agent harness on managed infrastructure. Claude can read files, run commands, browse and run code in a hosted sandbox; you call it through the Claude API.
| Free tier | No. Pay-as-you-go on the Claude API. |
|---|---|
| Paid plans | Model tokens at standard API rates plus $0.08 per session-hour, counted only while a session is running. Web search $10 per 1,000 searches. US-pinned inference is 1.1x. |
| Maker | Anthropic |
| Source | https://platform.claude.com/docs/en/about-claude/pricing |
| Last verified |
Building custom, stateful agents in code and tracing, evaluating and deploying them.
Orchestrating teams ("crews") of role-based agents in Python.
A lightweight, code-first way to build multi-agent workflows.
Visual workflow automation with AI agents, priced per workflow run, not per step.
No. Pay-as-you-go on the Claude API.
Model tokens at standard API rates plus $0.08 per session-hour, counted only while a session is running. Web search $10 per 1,000 searches. US-pinned inference is 1.1x.
Running long, asynchronous Claude agents without building your own agent loop or sandbox.
Claude Managed Agents is made by Anthropic.
Spotted something out of date? Tell us and we'll re-verify. Facts on this page come only from the linked sources.