Downloading 177GB Gated FLUX Model | Generated by AI

Home 2026.07 English 中文

---

hf-mirror.com is reachable! The model is 177 GB total. Here’s what I found:

---

📦 FLUX.2-dev — Model Info

Property	Value
Model	black-forest-labs/FLUX.2-dev
Pipeline	Flux2Pipeline (diffusers)
Total size	177 GB
Gated	✅ Yes — needs HF token + license acceptance
Disk space	2.9T available ✅
Mirror	hf-mirror.com (reachable ✅)

Files breakdown

• ae.safetensors — Autoencoder
• flux2-dev.safetensors — Main FLUX model
• text_encoder/model-00001~10-of-00010.safetensors — T5 text encoder (~10 shards)
• transformer/diffusion_pytorch_model-00001~07-of-00007.safetensors — Transformer (~7 shards)
• vae/diffusion_pytorch_model.safetensors — VAE

Issue: The model is gated

You need to:

1. Go to https://huggingface.co/black-forest-labs/FLUX.2-dev and accept the license (click “Agree”)
2. Get a HF token (read-only is enough) from https://huggingface.co/settings/tokens
3. Pass it to us so we can log in via huggingface-cli login --token YOUR_TOKEN

Do you have a HuggingFace token ready? If so, I’ll:

1. Set up a tmux session
2. Write the download script with HF_ENDPOINT=https://hf-mirror.com
3. Start the download to /root/FLUX.2-dev

---

Generated by AI. Curating and sharing still takes effort. If you find it useful, feel free to donate. WeChat: @lzwjava · X: @lzwjava · Say hi 👋

Back Donate