Blog
中文Swarms Blog
Engineering notes, product updates, and guides for building multi-agent systems with Swarms.
Subscribe via RSS
139 articles
Swarms API vs Amazon Bedrock Agents vs Vertex AI Agent Builder: Which Hosted Multi-Agent API Should You Pick in 2026?
A hands-on comparison of the three ways to run multi-agent systems without hosting them yourself: the Swarms API, Amazon Bedrock Agents and AgentCore, and Google's Vertex AI Agent Builder with ADK. Same task on each, with architectures, model choice, streaming, batch, pricing and time to first result.
How to Add a Multi-Agent Step to n8n, Zapier and Make with the Swarms API
Run a team of AI agents inside any no-code automation. One HTTP Request node, one JSON body, and the Swarms API returns the finished work. Step-by-step setup for n8n, Zapier and Make, plus three ready-to-paste recipes: lead enrichment, support triage and a weekly report.
Getting Started with the Swarms API MCP Server: Give Any AI Client the Whole Swarms API as Tools
A complete tutorial for the hosted Swarms MCP server at api.swarms.world/mcp. Connect Claude Code, Claude Desktop, Cursor or your own Python and TypeScript agents in two lines, learn the 23 tools, run your first agent and swarm from a chat prompt, read results and costs, and fix the errors you will hit on day one.
How Much Does It Cost to Run a Multi-Agent System? Real Numbers from the Swarms API
We ran the same task through a single agent and four multi-agent architectures on the Swarms API and read the bill from each response. Cost per run, cost per 1,000 runs, where the money goes, and the three levers that cut it: model mixing, max_loops and the night-time discount. With the benchmark script.
Build a Multi-Agent Backend in Node.js and TypeScript with the Swarms API
A complete TypeScript guide to the Swarms API: a typed fetch client, a Next.js route handler, a streaming route that shows each agent typing in the browser, a CLI script that runs on Node 24 with no build step, and a Vercel deploy. No Python required.
Build Your Own Deep Research API with Multi-Agent Swarms
A planner, parallel web-searching agents, a synthesizer and a fact-checking critic, wired together on the Swarms API in about 100 lines of Python and exposed as an HTTP endpoint with FastAPI. Real output, real costs, and a one-line cURL to try the research agent before you write any code.
AdvancedResearch: An Open-Source Implementation of Anthropic's Multi-Agent Research System
Anthropic published how its Research feature works: a lead agent plans, parallel subagents search, a citation agent checks every claim. AdvancedResearch is a 1,400-line Python package that implements that orchestrator-worker design on the Swarms API and Exa, with no agent framework. Here is how each idea from the paper maps to the code, a real planning run with its cost, and how to run it yourself.
Upgrading to Swarms v17: Breaking Changes and Migration Guide
Swarms v17 upgrade and migration guide: every removed class and parameter in InterLink, what to use instead, the changed defaults, and an upgrade checklist.
Using SQLite as Shared Memory for AI Agents
Shared memory for AI agents with SQLite: SqliteChannel gives agents and processes one durable, queryable conversation history. Store, share and run SQL over it.
Newsletter
Get the latest from Swarms
Build with Swarms Cloud or discover agents in the Marketplace. Sign up and get $5 in free API credits.