add README.md

README.md

@@ -0,0 +1,62 @@
+---
+license: cc-by-nc-4.0
+pipeline_tag: text-to-image
+library_name: pytorch
+tags:
+- diffusion-transformer
+- rectified-flow
+- text-to-image
+- from-scratch
+base_model:
+- black-forest-labs/FLUX.2-dev
+- google/flan-t5-base
+datasets:
+- LucasFang/FLUX-Reason-6M
+- bghira/photo-concept-bucket
+- detection-datasets/coco
+---
+
+# tinydit-256
+
+A 210M-parameter text-to-image diffusion transformer trained from scratch on one GPU (3.5 days, 400k steps,
+4.2M images at 256² area with aspect-ratio buckets). Frozen FLUX.2 autoencoder (32-channel latent) and frozen
+flan-t5-base text encoder; the transformer, recipe, data pipeline and evaluation are original.
+
+- Code, figures and every design decision with sources: https://github.com/ivanmikhnenkov/tinydit (tag `v1-pretrain`)
+- Try it: https://huggingface.co/spaces/ivanmikhnenkov/tinydit
+- Author: https://ivanmikhnenkov.com
+
+## Results (held-out prompts, 20 Euler steps, CFG 4, shift 2.8)
+
+| FID | FD-DINOv2 | object accuracy (COCO classes, Faster R-CNN) | CLIP | PickScore | HPSv2.1 |
+|---|---|---|---|---|---|
+| 27.0 | 218 | 90% | 0.322 | 20.9 | 0.254 |
+
+Strengths: single objects, animals, scenes, colours and simple relations, five aspect ratios. Known limits: readable
+text, close faces, crowds, counts above three.
+
+## Files
+
+- `model.safetensors`: EMA weights (bf16), step 400k
+- `config.json`: architecture and sampler defaults
+- `stats.json`: per-channel latent mean/std; latents are whitened with these after encoding and un-whitened before decoding
+
+## Usage
+
+Clone the GitHub repo, fetch the frozen FLUX.2 VAE (gated, needs your own HF token; it is not redistributed here)
+and flan-t5-base, then:
+
+```bash
+python -m tinydit.sample --ckpt model.safetensors --config run1 --sets NOVEL=prompts/novel.txt --out out.png --shape 320x208
+```
+
+or run the local playground (`python -m tinydit.playground`) for interactive prompts, sampling trajectories and
+attention maps. Architecture: adaLN-single DiT, dim 896 × 16 blocks, 2D RoPE, QK-norm, SwiGLU, 16 register tokens
+and 2 learned null key/value slots per cross-attention; rectified flow with a 2.8 timestep shift, cosine and
+dispersive auxiliary losses.
+
+## License
+
+Weights are released under CC BY-NC 4.0. They depend at inference on the FLUX.2 autoencoder (FLUX.2-dev
+non-commercial licence, gated) and were trained in part on images generated by FLUX.1-dev (FLUX-Reason-6M);
+use is non-commercial accordingly.