# deep-floyd/IF

Repository: https://github.com/deep-floyd/IF
Canonical: https://ross.abutalabs.com/products/if
Language: Python
License: NOASSERTION
License Family: other
Last push: 2024-04-14T05:26:35+00:00

## Health v2 (maintenance only)
Score: 22/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 0, release rhythm 8, longevity 94
- inputs: {"age_days": 1321, "days_push": 871, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_license
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 7804, forks 523 (observed 2026-08-28T04:10:05.151775+00:00)

## What it is
DeepFloyd IF is an open-source text-to-image model library implementing a cascaded pixel diffusion architecture with a frozen T5 text encoder and two super-resolution stages, producing photorealistic images up to 1024x1024 px. It ships as a Python package with notebooks and integrations for running the base and upscaler modules.

## Use cases
- generate photorealistic images from text prompts
- run a cascaded diffusion text-to-image pipeline locally
- upscale 64x64 generated images to 256px and 1024px
- research large UNet diffusion models with strong language understanding
- benchmark text-to-image models on COCO zero-shot FID

## When to choose
- you need high photorealism and strong text rendering in generated images
- you have 16-24GB of GPU VRAM available
- you want an open-source research-grade text-to-image model in Python

## When to avoid
- you lack a high-VRAM GPU
- you need actively maintained tooling or the latest model features
- you want a lightweight or CPU-only image generator

## Facets
- artifact type: library
- maturity: maintenance
- function: machine-learning, image-processing, deep-learning
- domain: artificial-intelligence, image-processing, deep-learning
- platform: python
- tags: text-to-image, diffusion-models, image-generation, super-resolution, t5-encoder, gpu, linux

## Member repositories
- deep-floyd/IF (main) score 22

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:10:05.151775+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 2026-08-29T17:34:37.511283+00:00, confidence not recorded.
  - readme: https://github.com/deep-floyd/IF (fetched 2026-08-28T04:10:05.151775+00:00, sha 0e4d8cdc7d6e)
- Data as of 2026-08-30T08:39:29.467469+00:00.