# ZhengYinan-AIR/Diffusion-Planner

[ICLR 2025 Oral] The official implementation of "Diffusion-Based Planning for Autonomous Driving with Flexible Guidance"

Repository: https://github.com/ZhengYinan-AIR/Diffusion-Planner
Canonical: https://ross.abutalabs.com/products/diffusion-planner
Homepage: https://zhengyinan-air.github.io/Diffusion-Planner/
Language: Python
License Family: other
Topics: autonomous-driving, diffusion-models, imitation-learning, motion-planning, pytorch
Last push: 2026-03-10T08:41:05+00:00

## Health v2 (maintenance only)
Score: 53/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 71, release rhythm 35, longevity 46
- inputs: {"age_days": 650, "days_push": 176, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1043, forks 174 (observed 2026-08-28T04:03:21.019060+00:00)

## What it is
Official PyTorch implementation of Diffusion Planner, an ICLR 2025 Oral paper using a DiT-based diffusion model for closed-loop autonomous driving motion planning with flexible classifier guidance. It jointly models prediction and planning as future trajectory generation and achieves real-time (~20Hz) inference on nuPlan without rule-based refinement.

## Use cases
- closed-loop motion planning for autonomous vehicles
- diffusion-based trajectory generation
- joint prediction and planning of vehicle behavior
- benchmarking learning-based planners on nuPlan
- research on classifier guidance for planning
- real-time driving policy inference

## When to choose
- researching diffusion models for autonomous driving planning
- evaluating learning-based planners in closed-loop nuPlan simulation
- exploring joint prediction-planning architectures
- building on published ICLR 2025 planning research

## When to avoid
- production autonomous driving systems without safety certification
- projects requiring a permissive license (no license provided)
- non-research users needing turnkey self-driving software
- planning tasks outside vehicle trajectory domains

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, deep-learning, simulation
- domain: autonomous-vehicles, machine-learning, deep-learning
- platform: python
- tags: diffusion-models, motion-planning, imitation-learning, nuplan, autonomous-driving, trajectory-prediction, research-code

## Member repositories
- ZhengYinan-AIR/Diffusion-Planner (main) score 53

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:21.019060+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 2026-08-30T07:02:29.134077+00:00, confidence not recorded.
  - readme: https://github.com/ZhengYinan-AIR/Diffusion-Planner (fetched 2026-08-28T04:03:21.019060+00:00, sha 832c2d1a720f)
  - homepage: https://zhengyinan-air.github.io/Diffusion-Planner/ (fetched 2026-08-29T13:03:41.756440+00:00, sha f6da9c4a36b7)
- Data as of 2026-08-30T08:39:29.467469+00:00.