# LuChengTHU/dpm-solver

Official code for "DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps" (Neurips 2022 Oral)

Repository: https://github.com/LuChengTHU/dpm-solver
Canonical: https://ross.abutalabs.com/products/dpm-solver
Language: Python
License: MIT
License Family: permissive
Topics: diffusion-models, machine-learning, score-based-generative-models, stable-diffusion
Last push: 2024-02-06T20:38:31+00:00

## Health v2 (maintenance only)
Score: 32/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 0, release rhythm 35, longevity 100
- inputs: {"age_days": 1487, "days_push": 939, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1852, forks 136 (observed 2026-08-28T04:05:44.418005+00:00)

## What it is
Official PyTorch implementation of DPM-Solver and DPM-Solver++, fast high-order ODE solvers for diffusion probabilistic model sampling that generate high-quality samples in 10-20 function evaluations without retraining. It works with both discrete-time and continuous-time diffusion models and is integrated into Hugging Face Diffusers.

## Use cases
- speed up stable diffusion image generation to 10-20 steps
- sample from diffusion models faster without retraining
- use a fast scheduler for text-to-image pipelines in diffusers
- accelerate guided diffusion model sampling
- reduce inference cost of diffusion model sampling

## When to choose
- you need fast, high-quality sampling from diffusion models with few function evaluations
- you use Stable Diffusion or guided diffusion and want a proven SOTA solver
- you want a solver that works without additional training

## When to avoid
- you need training or fine-tuning of diffusion models themselves
- your project requires non-diffusion generative models
- you need a maintained production library rather than research code

## Facets
- artifact type: library
- maturity: stable
- function: machine-learning, deep-learning, image-processing
- domain: machine-learning, deep-learning, image-processing, artificial-intelligence
- platform: python
- tags: diffusion-models, ode-solver, sampling, stable-diffusion, score-based-generative-models, research-code, gpu

## Member repositories
- LuChengTHU/dpm-solver (main) score 32

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:44.418005+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 2026-08-30T03:16:52.069931+00:00, confidence not recorded.
  - readme: https://github.com/LuChengTHU/dpm-solver (fetched 2026-08-28T04:05:44.418005+00:00, sha cd8f6d4702fe)
- Data as of 2026-08-30T08:39:29.467469+00:00.