Hiveflow
Visual workflow orchestration for AI agents and automation
Sim
Rapidly build, run, and scale AI agent workflows.
An open-source platform that lets developers and teams design, deploy, and manage AI agent workflows with a visual editor, local model support, and built-in orchestration.
Overview
Highlights
Visual low‑code flow editor for building agent pipelines
Plug‑and‑play local model support via Ollama (GPU or CPU)
Real‑time collaboration and socket‑based updates
Extensible background jobs and remote code execution
Pros
- Apache‑2.0 open‑source license
- Flexible deployment: cloud or self‑hosted
- Modern stack (Next.js, Bun, TypeScript)
- Built‑in vector embeddings for RAG
Considerations
- Requires Docker/Bun and PostgreSQL with pgvector
- GPU needed for optimal local model performance
- Learning curve for advanced orchestration features
- Limited pre‑built UI components for end users
Managed products teams compare with
When teams consider Sim, these hosted platforms usually appear on the same shortlist.
Looking for a hosted option? These are the services engineering teams benchmark against before choosing open source.
Fit guide
Great for
- Teams building custom AI assistants
- Enterprises needing on‑prem AI orchestration
- Developers seeking a reusable agent framework
- Organizations with strict data‑privacy requirements
Not ideal when
- Users without Docker or Linux environments
- Simple single‑turn chatbots without workflow needs
- Teams preferring Python‑centric ecosystems
- Small hobby projects lacking infrastructure
How teams use it
Customer support chatbot with knowledge base
Automates ticket triage using RAG and real‑time agent routing
Internal data analyst assistant
Executes SQL queries via remote code execution and returns insights in Slack
Automated content generation pipeline
Chains LLMs, image models, and post‑processing jobs to produce marketing assets
IoT device monitoring agent
Listens to sensor streams, triggers alerts, and runs corrective scripts autonomously
Tech snapshot
TypeScript74%
MDX25%
Python1%
CSS1%
Smarty1%
Shell1%
Frequently asked questions
What runtime does Sim use?
Sim runs on the Bun JavaScript runtime and is compatible with Node environments.
Do I need an external AI API?
No; you can run local models with Ollama or connect to any hosted LLM via API keys.
How is data stored?
PostgreSQL with the pgvector extension stores embeddings and workflow state.
Can I host Sim on my own servers?
Yes, via Docker Compose, the npm package, or VS Code dev containers; a cloud version is also available at sim.ai.
Is there built‑in authentication?
Sim uses Better Auth for secure user management out of the box.
Project at a glance
Active- Stars
- 25,801
- Watchers
- 25,801
- Forks
- 3,221
LicenseApache-2.0
Repo age1 year old
Last commityesterday
Primary languageTypeScript
Last synced yesterday