This is the FP16 conversion of HiDream-ai/HiDream-O1-Image-Dev — the distilled variant of HiDream-O1-Image — for use with ComfyUI. The Dev model runs in just 28 steps (half the steps of the full model) making it significantly faster while retaining strong output quality.
The Dev model uses a custom Euler scheduler with built-in noise scaling tuned for fewer steps. CFG is disabled — negative prompts have no effect in Dev mode.
BF16 vs FP16: Both use 16 bits per weight and have identical VRAM usage. BF16 has a wider dynamic range; FP16 has higher mantissa precision. Differences at inference are typically negligible. If you encounter NaN/Inf errors with FP16, switch to the BF16 variant.
The folder must contain the full Hugging Face support files alongside the weights:
config.json, chat_template.json, generation_config.json, preprocessor_config.json, tokenizer.json, tokenizer_config.json, vocab.json, merges.txt, model.safetensors
3. Load in ComfyUI
Use the workflow provided in the custom node repository. The loader will detect dev in the folder name and automatically apply Dev settings (28 steps, no CFG, Euler scheduler). Point the model loader to HiDream-O1-Image-Dev-fp16.
About HiDream-O1-Image
HiDream-O1-Image is a natively unified image generative foundation model built on a Pixel-level Unified Transformer (UiT) — no external VAEs, no disjoint text encoders. It encodes raw pixels, text, and task-specific conditions in a single shared token space, supporting:
At only 9B parameters it matches or exceeds much larger open-source DiTs and leading closed-source models. It debuted at #8 in the Artificial Analysis Text to Image Arena (2026-05-05).
Key Features
🧬 Pixel-Level Unified Transformer — end-to-end on raw pixels, no VAE, no disjoint text encoder
🎨 One Model, Many Tasks — T2I, editing, personalization, storyboard generation
⚡ 28-Step Distilled Dev — ~2× faster than the full model with minimal quality trade-off
🖼️ Native High Resolution — direct synthesis up to 2,048 × 2,048
