Can the RTX 5060 Ti 16 GB run FLUX.2 [klein] 4B?
Timed on my own card, not estimated: 2.9 seconds per 1024x1024 image with the FP8 file, peaking at 12.2 GB.
16-bit (7.8 GB) needs about 9.6 GB in use, so 6.4 GB of the 16 GB stays free: room for a larger image, a LoRA or a ControlNet.
01My measurements
| Label | Setup | Result | Peak VRAM | Date | Source |
|---|---|---|---|---|---|
| measured | flux-2-klein-4b-fp8.safetensors · 1024x1024 · 4 steps 3 timed runs after 1 warm-up (6.66 s incl. loading). Text encoder Qwen3 4B FP8. ComfyUI 0.25.0 portable, PyTorch 2.12 cu130, 32 GB RAM. 1024x1024, 4 steps, euler, Flux2Scheduler, C | 2.85 s / image | 12.2 GB | 2026-10-01 | |
| measured | flux-2-klein-4b.safetensors · 1024x1024 · 4 steps 3 timed runs after 1 warm-up (6.94 s incl. loading). Text encoder Qwen3 4B FP8. ComfyUI 0.25.0 portable, PyTorch 2.12 cu130, 32 GB RAM. 1024x1024, 4 steps, euler, Flux2Scheduler, C | 3.93 s / image | 14.1 GB | 2026-10-01 | |
| measured | flux-2-klein-4b-Q8_0.gguf · 1024x1024 · 4 steps 3 timed runs after 1 warm-up (8.95 s incl. loading). Text encoder Qwen3 4B FP8. ComfyUI 0.25.0 portable, PyTorch 2.12 cu130, 32 GB RAM. 1024x1024, 4 steps, euler, Flux2Scheduler, C | 4.34 s / image | 12.4 GB | 2026-10-01 |
02What to download
| Part | File | Folder | Size | |
|---|---|---|---|---|
| Model | flux-2-klein-4b.safetensors 16-bit · black-forest-labs/FLUX.2-klein-4B | models/diffusion_models | 7.8 GB | Download → |
| Text encoder | qwen_3_4b_fp8_mixed.safetensors Qwen3 4B FP8 · Comfy-Org/z_image_turbo | models/text_encoders | 5.6 GB | Download → |
| VAE | flux2-vae.safetensors FLUX.2 VAE · Comfy-Org/flux2-dev | models/vae | 0.3 GB | Download → |
| Total download · keep about the same free on disk | 13.7 GB | |||
Alternatives: Qwen3 4B BF16 (8.0 GB). Sizes from Hugging Face, 2026-10-01.
Free ComfyUI workflow, tested on my RTX 5060 Ti: FP8 workflow (.json). Drag the file into ComfyUI. All workflows →
03Every FLUX.2 klein 4B file on 16 GB
| File | Size | Needed | On this card | Quality | Download |
|---|---|---|---|---|---|
| 16-bit ← SAFETENSORS · black-forest-labs | 7.8 GB | 9.6 GB | fits · 6.4 GB spare | the original weights | Hugging Face → |
| FP8 SAFETENSORS · black-forest-labs | 4.1 GB | 5.9 GB | fits · 10.1 GB spare | practically identical to the original | Hugging Face → |
| Q8_0 GGUF · unsloth | 4.3 GB | 6.1 GB | fits · 9.9 GB spare | practically identical to the original | Hugging Face → |
| Q6_K GGUF · unsloth | 3.4 GB | 5.2 GB | fits · 10.8 GB spare | very close to the original | Hugging Face → |
| Q5_K_M GGUF · unsloth | 3.1 GB | 4.9 GB | fits · 11.1 GB spare | close; small differences in fine detail | Hugging Face → |
| Q4_K_M GGUF · unsloth | 2.6 GB | 4.4 GB | fits · 11.6 GB spare | good; some loss in fine detail and text | Hugging Face → |
| Q3_K_M GGUF · unsloth | 2.1 GB | 3.9 GB | fits · 12.1 GB spare | noticeable loss of detail | Hugging Face → |
| Q2_K GGUF · unsloth | 1.8 GB | 3.6 GB | fits · 12.4 GB spare | heavy loss; a last resort | Hugging Face → |
“Needed” adds 1 GB of working memory for FLUX.2 klein 4B and a 0.8 GB reserve. Sizes 2026-10-01.
04Questions
How much VRAM does FLUX.2 klein 4B need?
Around 3.6 GB with the smallest file (Q2_K) and 5.9 GB with an 8-bit file (FP8). The full 16-bit file needs about 9.6 GB.
Which FLUX.2 klein 4B file should I download for the RTX 5060 Ti 16 GB?
16-bit (7.8 GB) from black-forest-labs/FLUX.2-klein-4B.
FP8 or Q8_0 for FLUX.2 klein 4B on the RTX 5060 Ti 16 GB?
FP8 (4.1 GB): the RTX 5060 Ti 16 GB computes it natively, while Q8_0 (4.3 GB) is unpacked on the fly.
How much do I need to download for FLUX.2 klein 4B on the RTX 5060 Ti 16 GB?
13.7 GB: flux-2-klein-4b.safetensors, qwen_3_4b_fp8_mixed.safetensors, flux2-vae.safetensors.