<a target="_blank" rel="ugc" href="https://civarchive.com/models/1330309/chroma">Chroma by Lodestone, </a>quantized using GGUF by <a target="_blank" rel="ugc" href="https://huggingface.co/silveroxides/Chroma-GGUF">silveroxides</a> and <a target="_blank" rel="ugc" href="https://huggingface.co/Clybius/Chroma-GGUF">Clybius</a>.<h1 id="original-description:-bniolszjy">Original description:</h1>Chroma is a 8.9B parameter model based on FLUX.1-schnell (technical report coming soon!). It’s fully Apache 2.0 licensed, ensuring that anyone can use, modify, and build on top of it—no corporate gatekeeping.The model is still training right now, and I’d love to hear your thoughts! Your input and feedback are really appreciated.<h1 id="what-chroma-aims-to-do-q4hp0urt9">What Chroma Aims to Do</h1><ul><li>Training on a 5M dataset, curated from 20M samples including anime, furry, artistic stuff, and photos.</li><li>Fully uncensored, reintroducing missing anatomical concepts.</li><li>Built as a reliable open-source option for those who need it.</li></ul><h1 id="see-the-progress-fy3i5mmft">See the Progress</h1><ul><li>Hugging Face Repo: <a target="_blank" rel="ugc" href="https://huggingface.co/lodestones/Chroma">https://huggingface.co/lodestones/Chroma</a></li><li>Hugging Face Debug Repo: <a target="_blank" rel="ugc" href="https://huggingface.co/lodestones/chroma-debug-development-only">https://huggingface.co/lodestones/chroma-debug-development-only</a></li><li>Live WandB Training Logs: <a target="_blank" rel="ugc" href="https://wandb.ai/lodestone-rock/optimal%20transport%20unlocked">https://wandb.ai/lodestone-rock/optimal%20transport%20unlocked</a></li><li>ComfyUI Inference node [WIP]: <a target="_blank" rel="ugc" href="https://github.com/lodestone-rock/flux-mod">https://github.com/lodestone-rock/flux-mod</a></li><li>ComfyUI workflow: <a target="_blank" rel="ugc" href="https://huggingface.co/lodestones/Chroma/resolve/main/simple_workflow.json">https://huggingface.co/lodestones/Chroma/resolve/main/simple_workflow.json</a></li><li>Training code!: <a target="_blank" rel="ugc" href="https://github.com/lodestone-rock/flow">https://github.com/lodestone-rock/flow</a></li></ul><h1 id="support-open-source-ai-45grkye7x">Support Open-Source AI</h1>The current pretraining run has already used 5000+ H100 hours, and keeping this going long-term is expensive.If you believe in accessible, community-driven AI, any support would be greatly appreciated.👉 [<a target="_blank" rel="ugc" href="https://ko-fi.com/lodestonerock/goal?g=1">https://ko-fi.com/lodestonerock/goal?g=1</a>] — Every bit helps!ETH: 0x679C0C419E949d8f3515a255cE675A1c4D92A3d7their discord: <a target="_blank" rel="ugc" href="http://discord.gg/SQVcWVbqKx">discord.gg/SQVcWVbqKx</a>

Chroma GGUF (old versions)

One of the first versions that no longer requires CFG, so it should be set to 1, like when using Flux

chromaGGUFOldVersions_20250629CFGDistillQ8.gguf

2025-06-29-CFG-distill-Q8

v38-highres-Q8_0

v36-highres-fp8

v35-highres-Q8_0

v34-highres-Q8_0

v33-Q8_0

v32-Q8_0

v31-Q8_0

v30-Q8_0

v29.5-Q8

v29-fp8_scaled

v28-fp8_scaled

v28-Q6_K

v27_Q8_0

v27-Q4_K_M

v26-Q4_K_M

v25-fp8_e4m3fn_scaled

v24-Q4_K_M

v23-Q4_K_M

v22-Q4_K_M

v21-Q4_K_M

v20-Q4_K_M

v19-Q4_K_M

v18-Q4_K_M

v17-Q4_K_M

v15-Q4_K_M

v13-Q4_K_M

v12-fp8-e4m3fn

v12-Q4_K_M

v11-Q4_K_S

Files with SHA256: d31affcf6c18383f019c76d78f1bd6ae5d2e3a47c9639365356dc90896661dbc