Can the RX 9070 XT 16 GB run FLUX.1 Krea [dev]?
Yes, but only just: FP8 leaves 1.8 GB free on the RX 9070 XT 16 GB.
FP8 (11.9 GB) fits with only 1.8 GB to spare (about 14.2 GB of 16 GB).
01What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | flux1-krea-dev_fp8_scaled.safetensors FP8 · Comfy-Org/FLUX.1-Krea-dev_ComfyUI | models/diffusion_models | 11.9 GB | Download → |
| Text encoder | t5xxl_fp8_e4m3fn.safetensors T5-XXL FP8 · comfyanonymous/flux_text_encoders | models/text_encoders | 4.9 GB | Download → |
| Text encoder | clip_l.safetensors CLIP-L · comfyanonymous/flux_text_encoders | models/text_encoders | 0.2 GB | Download → |
| VAE | ae.safetensors FLUX.1 VAE (ae) · Comfy-Org/Lumina_Image_2.0_Repackaged | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 17.4 GB | |||
Alternatives: T5-XXL FP16 (9.8 GB). Sizes from Hugging Face, 2026-10-01.
02Every FLUX.1 Krea file on 16 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit SAFETENSORS · black-forest-labs | 23.8 GB | 26.1 GB | 10.1 GB too big | the original weights | Hugging Face → |
| FP8 ← SAFETENSORS · Comfy-Org | 11.9 GB | 14.2 GB | fits · 1.8 GB spare | practically identical to the original | Hugging Face → |
| Q8_0 GGUF · QuantStack | 12.7 GB | 15.0 GB | fits · 1.0 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · QuantStack | 9.8 GB | 12.1 GB | fits · 3.9 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · QuantStack | 8.4 GB | 10.7 GB | fits · 5.3 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · QuantStack | 6.9 GB | 9.2 GB | fits · 6.8 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · QuantStack | 5.4 GB | 7.7 GB | fits · 8.3 GB spare | noticeable loss of detail | Hugging Face → |
| Q2_K GGUF · QuantStack | 4.0 GB | 6.3 GB | fits · 9.7 GB spare | heavy loss; a last resort | Hugging Face → |
“Needed” adds 1.5 GB of working memory for FLUX.1 Krea and a 0.8 GB reserve. Sizes 2026-10-01.
03Measured and reported results
Nobody has sent measured numbers for this pair yet. If you run FLUX.1 Krea on a RX 9070 XT 16 GB, send your time per image and peak VRAM and it will appear here, credited.
04About the RX 9070 XT 16 GB
ComfyUI supports AMD GPUs on Windows officially since v0.7.0 (January 2026), through ROCm. Memory works the same as on NVIDIA, so the fit verdicts apply. Some custom nodes are NVIDIA-only. RX 9000 (RDNA 4) GPUs have FP8 hardware, and ComfyUI computes FP8 natively on them with PyTorch 2.7+ and ROCm 6.4+.
16 GB GDDR6 · 256-bit · 640 GB/s · RDNA 4. Everything that runs on it →
05Questions
How much VRAM does FLUX.1 Krea need?
Around 6.3 GB with the smallest file (Q2_K) and 14.2 GB with an 8-bit file (FP8). The full 16-bit file needs about 26.1 GB.
Which FLUX.1 Krea file should I download for the RX 9070 XT 16 GB?
FP8 (11.9 GB) from Comfy-Org/FLUX.1-Krea-dev_ComfyUI.
FP8 or Q8_0 for FLUX.1 Krea on the RX 9070 XT 16 GB?
FP8 (11.9 GB): the RX 9070 XT 16 GB computes it natively, while Q8_0 (12.7 GB) is unpacked on the fly.
How much do I need to download for FLUX.1 Krea on the RX 9070 XT 16 GB?
17.4 GB: flux1-krea-dev_fp8_scaled.safetensors, t5xxl_fp8_e4m3fn.safetensors, clip_l.safetensors, ae.safetensors.