FLUX.2 [klein] 9B is an image generation model by Black Forest Labs (9B; text encoder: Qwen3-8B). Higher quality than the 4B; officially about 29 GB of VRAM (RTX 4090 and up).
Change precision, resolution →
| Precision | Model size | 1024×1024 | 1536×1536 |
|---|---|---|---|
| BF16 / FP16 (uncompressed) | 18.0 GB | 19.3 GB all in VRAM 28.1 GB | 20.2 GB all in VRAM 29.0 GB |
| FP8 (half size, near-lossless) | 9.2 GB | 10.4 GB all in VRAM 19.3 GB | 11.3 GB all in VRAM 20.2 GB |
| GGUF Q8_0 (similar to FP8) | 9.6 GB | 10.9 GB all in VRAM 19.7 GB | 11.8 GB all in VRAM 20.6 GB |
| GGUF Q4_K_S (popular low-VRAM choice) | 5.4 GB | 8.9 GB all in VRAM 15.5 GB | 8.9 GB all in VRAM 16.4 GB |
Bold = VRAM needed when ComfyUI swaps the text encoder and the model. "All in VRAM" = keeping both loaded. Text encoder assumed at FP8. Official: about 29 GB of VRAM (RTX 4090 and up) (measured with the official implementation).
| GPU | VRAM | Best precision | VRAM needed | Verdict | Buy |
|---|---|---|---|---|---|
| RTX 3080 (10GB) previous gen | 10 GB | GGUF Q4_K_S | 8.9 GB | ○ Runs with model swapping | Check price |
| RTX 2080 Ti previous gen | 11 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 3060 12GB previous gen | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 6700 XT previous gen | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Intel Arc B580 | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4070 Ti previous gen | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4070 / SUPER previous gen | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 5070 | 12 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4060 Ti 16GB previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 7600 XT previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 9060 XT 16GB | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 5060 Ti 16GB | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 6800 previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 7800 XT previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 9070 XT | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 9070 | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4070 Ti SUPER previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4080 previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 4080 SUPER previous gen | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 5070 Ti | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| RTX 5080 | 16 GB | FP8 | 10.4 GB | ○ Runs with model swapping | Check price |
| Radeon RX 7900 XT previous gen | 20 GB | BF16 / FP16 | 19.3 GB | ○ Runs with model swapping | Check price |
| RTX 3090 previous gen | 24 GB | BF16 / FP16 | 19.3 GB | ○ Runs with model swapping | Check price |
| Radeon RX 7900 XTX previous gen | 24 GB | BF16 / FP16 | 19.3 GB | ○ Runs with model swapping | Check price |
| RTX 4090 previous gen | 24 GB | BF16 / FP16 | 19.3 GB | ○ Runs with model swapping | Check price |
| RTX 5090 | 32 GB | BF16 / FP16 | 19.3 GB | ◎ Fits entirely in VRAM | Check price |
ComfyUI also runs on AMD Radeon (ROCm) and Intel Arc, but setup is harder than on NVIDIA and some features (such as FP8 acceleration) are not supported.
→ Image & Video AI VRAM Calculator (all 15 models) / Local LLM VRAM Calculator
Estimates for ComfyUI, about ±20%. Model specs from the publisher's config files and released file sizes (checked October 2026).