# Chameleon: Dynamic Format Adapter for Efficient Diffusion

arXiv: https://arxiv.org/abs/2609.33496 · HTML: https://mlsanity.com/paper/2609.33496/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Placed in Diffusion foundations and sampling, one of 14 directions in Diffusion models.
- Builds on 28 of its 34 references that are in the map.
- First posted on arXiv 27 Sep 2026.

## Placed in

- Diffusion models / [Diffusion foundations and sampling](https://mlsanity.com/diffusion/diffusion-foundations-and-sampling/)

## Builds on (most cited first, 28 of 28)

- [Learning Transferable Visual Models From Natural Language Supervision](https://mlsanity.com/paper/2103.00020/) (2021)
- [Microsoft COCO: Common Objects in Context](https://mlsanity.com/paper/1405.0312/) (2014)
- [SDXL: Improving Latent Diffusion Models for High-Resolution Image Synthesis](https://mlsanity.com/paper/2307.01952/) (2023)
- [DPM-Solver: A Fast ODE Solver for Diffusion Probabilistic Model Sampling in Around 10 Steps](https://mlsanity.com/paper/2206.00927/) (2022)
- [SmoothQuant: Accurate and Efficient Post-Training Quantization for Large Language Models](https://mlsanity.com/paper/2211.10438/) (2022)
- [AWQ: Activation-aware Weight Quantization for LLM Compression and Acceleration](https://mlsanity.com/paper/2306.00978/) (2023)
- [PixArt-α: Fast Training of Diffusion Transformer for Photorealistic Text-to-Image Synthesis](https://mlsanity.com/paper/2310.00426/) (2023)
- [Latent Consistency Models: Synthesizing High-Resolution Images with Few-Step Inference](https://mlsanity.com/paper/2310.04378/) (2023)
- [Adversarial Diffusion Distillation](https://mlsanity.com/paper/2311.17042/) (2023)
- [A White Paper on Neural Network Quantization](https://mlsanity.com/paper/2106.08295/) (2021)
- [Q-Diffusion: Quantizing Diffusion Models](https://mlsanity.com/paper/2302.04304/) (2023)
- [Post-training Quantization on Diffusion Models](https://mlsanity.com/paper/2211.15736/) (2022)
- [PTQD: Accurate Post-Training Quantization for Diffusion Models](https://mlsanity.com/paper/2305.10657/) (2023)
- [ViDiT-Q: Efficient and Accurate Quantization of Diffusion Transformers for Image and Video Generation](https://mlsanity.com/paper/2406.02540/) (2024)
- [Q-DiT: Accurate Post-Training Quantization for Diffusion Transformers](https://mlsanity.com/paper/2406.17343/) (2024)
- [FP8 Quantization: The Power of the Exponent](https://mlsanity.com/paper/2208.09225/) (2022)
- [Temporal Dynamic Quantization for Diffusion Models](https://mlsanity.com/paper/2306.02316/) (2023)
- [PTQ4DiT: Post-training Quantization for Diffusion Transformers](https://mlsanity.com/paper/2405.16005/) (2024)
- [EfficientDM: Efficient Quantization-Aware Fine-Tuning of Low-Bit Diffusion Models](https://mlsanity.com/paper/2310.03270/) (2023)
- [MixDQ: Memory-Efficient Few-Step Text-to-Image Diffusion Models with Metric-Decoupled Mixed Precision Quantization](https://mlsanity.com/paper/2405.17873/) (2024)
- [Rethinking floating point for deep learning](https://mlsanity.com/paper/1811.01721/) (2018)
- [Integer or Floating Point? New Outlooks for Low-Bit Quantization on Large Language Models](https://mlsanity.com/paper/2305.12356/) (2023)
- [BinaryDM: Accurate Weight Binarization for Efficient Diffusion Models](https://mlsanity.com/paper/2404.05662/) (2024)
- [Neural Network Training with Approximate Logarithmic Computations](https://mlsanity.com/paper/1910.09876/) (2019)
- [FP4DiT: Towards Effective Floating Point Quantization for Diffusion Transformers](https://mlsanity.com/paper/2503.15465/) (2025)
- [Q-Sched: Pushing the Boundaries of Few-Step Diffusion Models with Quantization-Aware Scheduling](https://mlsanity.com/paper/2509.01624/) (2025)
- [OASIS: Optimized Lightweight Autoencoder System for Distributed In-Sensor computing](https://mlsanity.com/paper/2505.02256/) (2025)
- [EntroLLM: Entropy Encoded Weight Compression for Efficient Large Language Model Inference on Edge Devices](https://mlsanity.com/paper/2505.02380/) (2025)

## Abstract (arXiv)

Post-training quantization (PTQ) is the standard way to run modern diffusion models on memory-constrained accelerators, yet every existing diffusion PTQ scheme fixes the mathitnumber format in advance and only tunes the scale, zero point, or per-layer bit-width. At a fixed bit-width the best format depends on the distribution being encoded, and that distribution differs across weight channels, across layers, and along the diffusion timestep, where activation distributions slide from heavy-tailed and noise-dominated to tightly clustered and structured. We propose Chameleon, a PTQ framework that holds the bit-width fixed and treats the format itself as a discrete variable, chosen per weight channel and per (layer, timestep bucket) activation tensor. Activation formats come from INT8, FP8 E4M3, FP8 E5M2, MXFP8, MXINT8, selected ahead of time from two cheap statistics (empirical kurtosis and the closed-form diffusion SNR) and stored in a lookup table; weight formats come from INT8, MXINT8 at 8 bits or INT4, NF4, FP4 E2M1, MXINT4, MXFP4 at 4 bits, selected offline by reconstruction error. An architectural fork adapts the same selection layer to multi-step UNets, single-step distilled models, and Diffusion Transformers. Across SDXL, SDXL-Turbo, and PixArt-α on COCO-2014, Chameleon achieves the best FID in all six backbone × bit-width settings, with CLIP within 0.24 of the FP16 reference and the best of all quantized methods at W4A8.