Agent Orchestration Framework
Orchflow
A dependency-free Python framework for readable multi-agent pipelines: sequential, parallel, conditional, and resumable flows.
0 deps
zero runtime dependencies in core
Orchflow live demo
live Flow.events() stream
Runtime graph
Ready to run.
Final
No output yet.
Inspector
No event selected.
Video walkthrough
Watch the complete Orchflow demo
A six-minute walkthrough of the same project story: readable Python steps, parallel branches, live events, retries, traces, and JSON checkpoint resume.
What it proves
A live case for readable multi-agent workflows
Readable orchestration
The running demo is still a normal Flow([...]) list: plan, a parallel group, synthesis, condition, finalizer.
Parallel work you can inspect
The three research agents share a parallel_group_id, so the UI can prove they fanned out and rejoined.
Resume is visible
Failure mode saves a JSON checkpoint, reloads it, and appends new traces after resume.
Safe live model demo
Visitors can adjust small inputs and model presets, while the backend enforces allowlisted models and rate limits.
The API
The backend is still plain Python
The portfolio UI is only a viewer. The value comes from Orchflow: step functions stay readable, while the framework handles fan-out, retries, lifecycle events, checkpoints, and resume.
from orchflow import Agent, Flow, StepContext, condition, step@step(name="technical_research", retry=2)async def technical_research(input: dict, context: StepContext):return await researcher.run(prompt, context=context)
flow = Flow([plan,[market_research, technical_research, risk_review],synthesize,condition(when=lambda ctx: ctx.previous["quality_score"] >= 0.8,then=publish_ready,otherwise=revise,),finalize,])
Get started
Install from PyPI
$ pip install orchflow$ pip install "orchflow[litellm]"
v0.5.0 · tag-based PyPI releases · core has zero runtime dependencies.