This is a GGUF quantized version of <a target="_blank" rel="ugc" href="https://huggingface.co/black-forest-labs/FLUX.2-klein-4B">FLUX.2-klein-4B</a>. unsloth/FLUX.2-klein-4B-GGUF uses <a target="_blank" rel="ugc" href="https://docs.unsloth.ai/basics/unsloth-dynamic-2.0-ggufs">Unsloth Dynamic 2.0</a> methodology for SOTA performance.<ul><li>Important layers are upcasted to higher precision.</li><li>Uses tooling from <a target="_blank" rel="ugc" href="https://github.com/city96/ComfyUI-GGUF">ComfyUI-GGUF</a> by city96.</li></ul>

Unsloth - FLUX.2-Klein-4B-GGUF (Distilled)

unslothFLUX2Klein4BGGUF_q80.gguf

Q8_0

FP16

BF16

Q6_K

Q5_K_S

Q5_K_M

Q5_1

Q5_0

Q4_K_M

Q4_1

Q4_0

Q4_K_S

Q3_K_M

Q3_K_S

Q2_K

oryginal repo: <a target="_blank" rel="ugc" href="https://huggingface.co/unsloth/FLUX.2-klein-4B-GGUF">https://huggingface.co/unsloth/FLUX.2-klein-4B-GGUF</a>plus simple workflowcfg scale: I use a range from 1 to 1.2

Flux2Klein4B -Q8_0

flux2klein4bQ80_v10.gguf

flux2klein4bQ80_v10.json

v1.0

Files with SHA256: 24a812e8f9b640e21c784164ea48571d8017ca22873696f4badeeabb006d509c