sd15-geoflow-characters

AbstractPhil/sd15-geoflow-characters

sd15-geoflow-characters is a text-to-image model from AbstractPhil. Use it when you need an image from a text prompt. It is set up for sd15-flow-trainer. The card lists the license as mit.

Geometric cross-attention prior for SD1.5 using pentachoron (4-simplex) structures.

Downloads · 30 days

Access

Public

Updated Feb 7, 2026

Repo size

33.1 MB

Likes

Public

Hugging Face

Repo makeup

Click a slice to open those files.

.safetensors19.4 MB · 58%

At a glance

Task: Text-to-Image
Library: sd15-flow-trainer
License: mit
Access: Public
Created: Feb 7, 2026
Updated: Feb 7, 2026
SHA: b299e7ba

Base models

Model card

From the Hugging Face model README

KSimplex Geometric Attention Prior

Geometric cross-attention prior for SD1.5 using pentachoron (4-simplex) structures.

Before and After

Pretrain

Final

Architecture

Component	Params
SD1.5 UNet (frozen)	859,520,964
Geo prior (trained)	4,845,725

The geometric prior modulates CLIP encoder hidden states through 4-layer stacked k-simplex attention before they reach the 16 cross-attention blocks in the UNet.

Simplex Configuration

Parameter	Value
k (simplex dim)	4
Embedding dim	32
Feature dim	768
Stacked layers	4
Attention heads	8
Base deformation	0.25
Residual blend	learnable
Timestep conditioned	True

Usage

from sd15_trainer_geo.pipeline import load_pipeline, load_geo_from_hub

# Load base SD1.5 + fresh geo prior
pipe = load_pipeline()

# Load trained geo weights from this repo
load_geo_from_hub(pipe, "AbstractPhil/sd15-geoflow-characters")

# Or one-shot: load base + geo in one call
pipe = load_pipeline(geo_repo_id="AbstractPhil/sd15-geoflow-characters")