# cumulo-autumn/StreamDiffusion

StreamDiffusion: A Pipeline-Level Solution for Real-Time Interactive Generation

Repository: https://github.com/cumulo-autumn/StreamDiffusion
Canonical: https://ross.abutalabs.com/products/streamdiffusion
Language: Python
License: Apache-2.0
License Family: permissive
Last push: 2024-12-04T08:05:55+00:00

## Health v2 (maintenance only)
Score: 17/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 72
- inputs: {"age_days": 1009, "days_push": 637, "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 10806, forks 834 (observed 2026-08-28T04:10:44.366083+00:00)

## What it is
StreamDiffusion is a Python pipeline for real-time interactive diffusion-based image generation, achieving 100+ fps on modern GPUs. It optimizes the standard Stable Diffusion pipeline with stream batching, residual classifier-free guidance, similarity filtering, and KV-cache pre-computation.

## Use cases
- generate images in real time from text prompts
- run interactive img2img generation from a webcam feed
- build a live visual effect tool with stable diffusion
- speed up stable diffusion inference to 100 fps
- create real-time AI art installations

## When to choose
- you need low-latency, high-fps diffusion image generation
- you want to build interactive real-time AI image applications on a strong GPU
- you want an optimized alternative to the standard diffusers pipeline

## When to avoid
- you lack a modern NVIDIA GPU
- you need maximum image quality over speed
- you need a general-purpose diffusion library rather than a real-time pipeline

## Facets
- artifact type: library
- maturity: active
- function: image-processing, machine-learning, deep-learning, gpu-computing, stable-diffusion
- domain: image-processing, deep-learning, gpu-computing
- platform: python, windows
- tags: diffusion-pipeline, real-time-generation, stable-diffusion, txt2img, img2img, pytorch, real-time, gpu, linux, macos

## Member repositories
- cumulo-autumn/StreamDiffusion (main) score 17

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:44.366083+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 2026-08-29T17:17:39.552959+00:00, confidence not recorded.
  - readme: https://github.com/cumulo-autumn/StreamDiffusion (fetched 2026-08-28T04:10:44.366083+00:00, sha b326930d9dbe)
  - registry_pypi: https://pypi.org/pypi/streamdiffusion/json (fetched 2026-08-29T08:16:35.765230+00:00, sha a3afd6aedb60)
- Data as of 2026-08-30T08:39:29.467469+00:00.