# flow-project/flow

Computational framework for reinforcement learning in traffic control

Repository: https://github.com/flow-project/flow
Canonical: https://ross.abutalabs.com/products/flow-project-flow
Language: Python
License: MIT
License Family: permissive
Topics: reinforcement-learning, traffic-control, benchmark, autonomous, vehicle-control, sumo
Last push: 2024-07-27T08:31:17+00:00

## Health v2 (maintenance only)
Score: 23/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 8, longevity 100
- inputs: {"age_days": 3297, "days_push": 767, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1188, forks 395 (observed 2026-08-28T04:03:55.315514+00:00)

## What it is
Flow is a computational framework for deep reinforcement learning and control experiments in traffic microsimulation, built on top of the SUMO simulator. It provides benchmarks and tools for studying mixed-autonomy traffic scenarios such as vehicle control and traffic flow optimization.

## Use cases
- train reinforcement learning agents to control traffic flow
- benchmark RL algorithms on traffic control tasks
- simulate mixed-autonomy traffic with SUMO
- research autonomous vehicle control in traffic
- run traffic microsimulation experiments for academic papers
- study ramp metering and intersection control with RL

## When to choose
- you need RL environments for traffic control research
- you want reproducible benchmarks for mixed-autonomy traffic
- you need to couple deep RL libraries with the SUMO traffic simulator

## When to avoid
- you need general-purpose robotics or game RL environments unrelated to traffic
- you need a production traffic management system rather than a research framework
- you require active commercial support or frequent updates

## Facets
- artifact type: framework
- maturity: maintenance
- function: reinforcement-learning, simulation, benchmarking, machine-learning
- domain: reinforcement-learning, autonomous-vehicles, simulation, machine-learning
- platform: python
- tags: traffic-microsimulation, sumo, mixed-autonomy-traffic, deep-rl, traffic-control, benchmark-suite, research, linux, macos

## Member repositories
- flow-project/flow (main) score 23

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:55.315514+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T06:23:36.452615+00:00, confidence not recorded.
  - readme: https://github.com/flow-project/flow (fetched 2026-08-28T04:03:55.315514+00:00, sha 8489019ff8ce)
- Data as of 2026-08-30T08:39:29.467469+00:00.