<p><span style="color:rgb(193, 194, 197)">Flux1 Depth Dev  converted to fp8_e4m3fn to reduce memory consumption (x0.5)</span></p>

Flux1 Depth Dev fp8 e4m3fn

<p>Flux1 Dev Depth converted to fp8_e4m3fn to reduce memory consumption (x0.5)</p>

flux1DepthDevFp8_v10.safetensors

v1.0

<h1 id="flux1-depth-dev-fp8-controlnet-model-by-bfl-wbll7eyo9">flux1-depth-dev-fp8 ControlNet model by BFL</h1><h2 id="key-features-zv26mm9uw">Key Features</h2><p></p><ol><li><p>Cutting-edge output quality.</p></li><li><p>It blends impressive prompt adherence with maintaining the structure of source images based on depth maps.</p></li><li><p>Trained using guidance distillation, making <code>FLUX.1 depth [dev]</code> more efficient.</p></li><li><p>Open weights to drive new scientific research, and empower artists to develop innovative workflows.</p></li><li><p>Generated outputs can be used for personal, scientific, and commercial purposes as described in the <code>FLUX.1 [dev]</code> Non-Commercial License.</p><p></p></li></ol><h2 id="usage-ef156i3fr">Usage</h2><p>We provide a reference implementation of <code>FLUX.1 Canny [dev]</code>, as well as sampling code, in a dedicated <a target="_blank" rel="ugc" href="https://github.com/black-forest-labs/flux"><strong><u>github repository</u></strong></a>. Developers and creatives looking to build on top of <code>FLUX.1 depth [dev]</code> are encouraged to use this as a starting point.</p>

Files with SHA256: 0b0c3411700edf63ba5445b08b4627a44a45d0f946dbd63bf89281ccd3c1d010