What GPU do I need to run black-forest-labs/FLUX.2-klein-base-4B?
3.9B parameters, published in BF16. View on Hugging Face
FLUX.2-klein-base-4B is published by black-forest-labs on Hugging Face, with 278,753 downloads and 158 likes to date. It's a unlisted-architecture model built for image-to-image, published natively in BF16.
VRAM required & cheapest live GPU fit
Required VRAM = weight size at each precision times a flat 1.2 overhead for activations and fragmentation. The KV-cache grows with context and is not in that factor; it is listed per model below where the architecture is published. Full formula and assumptions: methodology.
| Precision | Weight size | Required VRAM | Cheapest live fit | GPUs needed | Est. $/hr (full fit) |
|---|---|---|---|---|---|
| BF16 | 7.2 GB | 8.7 GB | RTX 5060 Ti | 1 | $0.145/hr |
| FP8 (quantized) | 3.6 GB | 4.3 GB | RTX 5060 Ti | 1 | $0.145/hr |
| INT4 (quantized) | 1.8 GB | 2.2 GB | RTX 3070 | 1 | $0.088/hr |
A GPU is only matched to a row if its hardware supports that precision, and the primary recommendation is always a single-GPU fit when one exists.
INT4 requires a quantized checkpoint actually published for this model, check its Hugging Face page before relying on this row.
Cheapest way to run FLUX.2-klein-base-4B at its published (BF16) precision: 1× RTX 5060 Ti, at $0.145/hr per GPU ($0.145/hr total). Quantizing to FP8 or INT4 (rows above) can cost less, but requires a compatible quantized checkpoint to exist for this model.
FLUX.2-klein-base-4B: common questions
Does FLUX.2-klein-base-4B fit on a 12 GB GPU?
Yes. At BF16 it needs 8.7 GB of VRAM, so a 12 GB card holds it with 3.3 GB to spare. An 8 GB card is not enough for it at BF16.
What is the least VRAM FLUX.2-klein-base-4B can run in?
2.2 GB, at INT4 (quantized), which fits a 6 GB card, against 8.7 GB at BF16. That row assumes an INT4 checkpoint has actually been published for this model, so check its Hugging Face page before planning around it.
Does quantizing FLUX.2-klein-base-4B lower the GPU bill?
Yes. At BF16 the cheapest live fit is one RTX 5060 Ti at $0.145/hr. At INT4 (quantized) it drops to one RTX 3070 at $0.088/hr, provided a quantized checkpoint exists for it.
Weight-to-VRAM math, the fit rules, and how live prices are normalized: full methodology.
More FLUX.2 models
- FLUX.2-klein-4B (3.9B, BF16)
- flux2-klein-9b-uncensored-text-encoder (8.2B, F16)
- FLUX.2-klein-9B (9.1B, BF16)
- BluePixel-Turbo (9.1B, F32)
- FLUX.2-dev (32.2B, BF16)
Alternatives at this size
Other models for image-to-image within about a third of FLUX.2-klein-base-4B's 3.9B parameters, from other model lines.
- flux-3-image-klein (5.5B, BF16)
More on FLUX.2-klein-base-4B
Fits on a 12 GB GPU at BF16: every model that fits in 12 GB.
Fits on an 8 GB GPU at FP8: every model that fits in 8 GB.
Related reading: H100 pricing and specs, Fast diffusion inference on GPU VMs, and What AI inference is.
ComfyUI on our compute
Rent a GPU with ComfyUI already installed. Your custom nodes and models are there next time.
$29 a month, or pay as you go