FLUX.1 [dev] from Black Forest Labs (ex-Stability researchers) is the highest-quality open image model — beats SDXL and approaches Midjourney v6 on photorealism and prompt following. The 12 B parameter footprint demands ~24 GB VRAM at FP16, ~7 GB at Q4 (NF4) using community quants. The non-commercial license is the main adoption blocker; FLUX.1 [schnell] is the Apache-licensed alternative.
FLUX.1 [dev] needs at minimum 7.3 GB of VRAM at Q4_K_M quantization (26.4 GB at FP16). The cheapest GPU that comfortably fits with KV-cache headroom is the Intel Arc B570 (10 GB VRAM, $219 MSRP). Community benchmark submissions are open. This model fits a single consumer GPU under 48 GB, so a one-card build works.
Weights only. Add ~20-30% for KV cache at typical context lengths.
| Quant | VRAM | Quality | Notes |
|---|---|---|---|
| FP16 | 26.4 GB | Reference | Training-precision weights |
| Q8_0 | 13.2 GB | Near-lossless | 8-bit, ~0.1% perplexity hit |
| Q6_K | 10.7 GB | Very high | 6-bit, near Q8 quality |
| Q5_K_M | 9.1 GB | High | Strong middle ground |
| Q4_K_M | 7.3 GB | Balanced (recommended) | Default for local deployments |
| Q4_0 | 7.4 GB | Legacy 4-bit | Older GGUF, kept for compatibility |
| Q3_K_M | 5.7 GB | Lossy | When VRAM is very tight |
| Q2_K | 4.2 GB | Extreme | Rescue option, quality degrades visibly |
Filtered from MyAIHardware's GPU database. FP16 needs 26.4 GB; Q4_K_M needs 7.3 GB.
We don't have community benchmarks for this exact model yet.
Official weights + popular runtime tags.
Step-by-step guides that use this model.
Models in a similar size or capability class.
Use these in papers, blog posts, or internal docs.
@misc{flux-1-dev-2024,
title={FLUX.1 [dev]},
author={Black Forest Labs},
year={2024},
url={https://huggingface.co/black-forest-labs/FLUX.1-dev}
}Black Forest Labs (2024). FLUX.1 [dev] [Model card]. Hugging Face. https://huggingface.co/black-forest-labs/FLUX.1-dev
FLUX.1 [dev] (FLUX, Black Forest Labs, 2024). Available at https://huggingface.co/black-forest-labs/FLUX.1-dev.