# Hugging Face — Efficient Controllable Generation for SDXL with T2I-Adapters

- Company: Hugging Face (huggingface.co)
- Announced: 2023-09-08
- Category: open-weight-release
- Coverage: not counted
- Announcement: yes
- Group: models
- Source: https://huggingface.co/blog/t2i-sdxl-adapters
- Record: https://forck.live/items/2001-efficient-controllable-generation-for-sdxl-with-t2i-adapters
- Subject: Platform
- Open weights: yes
- Models affected: T2I-Adapter-SDXL, SDXL

Hugging Face and the T2I-Adapter authors release T2I-Adapter checkpoints for Stable Diffusion XL, enabling efficient controllable generation with smaller and faster adapters compared to ControlNet, supporting conditions like sketch, canny, lineart, depth, and openpose.

## Evidence

Verbatim from https://huggingface.co/blog/t2i-sdxl-adapters:

> we share our findings from training T2I-Adapters on SDXL from scratch, some appealing results, and, of course, the T2I-Adapter checkpoints on various conditionings (sketch, canny, lineart, depth, and openpose)!

---

Record: https://forck.live/items/2001-efficient-controllable-generation-for-sdxl-with-t2i-adapters
Catalogue: https://forck.live/llms.txt
Current issue: https://forck.live/feed.md