DiffusionBench: Towards Holistic Evaluation of Generative Diffusion Transformers

##############################################################################
##############################################################################

📣 Announcement post:[Call for DiffusionBench: A Holistic Benchmark for Diffusion Transformers]. Help us grow the benchmark with new evaluation axes, new metrics, and faithful reproductions of published methods.

This repo contains the unified codebase for DiffusionBench. It supports training and evaluation across different generation tasks (ImageNet, T2I, ...) through a single interface. Please see the sections below for the detailed structure. Come join us!

Text-to-image samples at 256×256 from models trained for 200K iterations using DiffusionBench.

curl -LsSf https://astral.sh/uv/install.sh | sh

uv sync

uv run python scripts/prepare.py --data {all,imagenet,t2i,eval}

uv run hf download diffusion-bench/diffusion-bench --local-dir pretrained_models --exclude .gitattributes

Reproduction flow: Stage 1 → Stage 2. Set these environment variables first (used for the output directory and W&B logging):

export EXPERIMENT_NAME=<run-name>
export ENTITY=<wandb-entity>
export PROJECT=<wandb-project>
export WANDB_KEY=<key>

Stage 1. Train the RAE tokenizer:

uv run torchrun --standalone --nproc_per_node=8 \
    src/train_stage1.py \
    --config [STAGE1_CONFIG_PATH] \
    --results-dir results/stage1 --precision bf16 --compile --wandb

Stage 2. Train the diffusion model on VAE/RAE/Pixel space:

uv run torchrun --standalone --nproc_per_node=8 \
    src/train.py \
    --config [STAGE2_CONFIG_PATH] \
    --results-dir results/stage2 --precision bf16 --compile --wandb

Stage 2 training configs run online evaluation during training (the eval:

block). For standalone evaluation of a released checkpoint, use the ** sampling/** configs — each embeds

stage_2.ckpt

(pointing into pretrained_models/

) and the eval-time guidance, so the weights load automatically:

export EXPERIMENT_NAME=<run-name>

uv run torchrun --nproc_per_node=8 src/offline_eval_stage1.py --config [STAGE1_CONFIG_PATH]

uv run torchrun --nproc_per_node=8 src/offline_eval.py --config [STAGE2_CONFIG_PATH]
configs/
├── stage1/
└── stage2/
    ├── training/
    │   ├── imagenet/
    │   └── t2i/
    └── sampling/
        ├── imagenet/
        └── t2i/

Stage 2 spans VAE (11), RAE (6), REG (4), and Pixel (3) families, identical across ImageNet and T2I. Swap any config between tasks with a single path change. The sampling/

set mirrors training/

but adds the trained checkpoint and eval-time guidance, so it runs offline eval directly.

For ImageNet, pick the CFG-off baseline ([STAGE2_CONFIG_PATH].yaml

) or the per-model best-CFG variant ([STAGE2_CONFIG_PATH]-cfg<scale>-t0.0-0.9.yaml

).

AutoResearch We welcome contributions! Please refer to docs/contributors.md and

for further details.

docs/contributing.md

The codebase is built upon some amazing projects:

We thank the authors for making their work publicly available.