About
FLUX.2 Klein 4B — the fastest model in the FLUX family.
Unifies text-to-image and image editing in one compact architecture.
Two variants: Base (undistilled, best for fine-tuning) and Distilled (4-step, ~1.2s on 5090).
Only 8.4 GB VRAM for distilled variant. Supports style transforms, semantic edits,
object replacement/removal, multi-reference composition, and iterative edits.
Variants
| Variant | Format | Size | VRAM | Install |
|---|---|---|---|---|
| distilled-fp8 ⓘ | safetensors | 4.5 GB | 8+ GB | mods install flux2-klein-4b --variant distilled-fp8 |
| base-fp8 ⓘ | safetensors | 4.5 GB | 10+ GB | mods install flux2-klein-4b --variant base-fp8 |
| bf16 ⓘ | gguf | 7.8 GB | 10+ GB | mods install flux2-klein-4b --variant bf16 |
| gguf-q8-0 ⓘ | gguf | 4.3 GB | 6+ GB | mods install flux2-klein-4b --variant gguf-q8-0 |
| gguf-q6-k ⓘ | gguf | 3.4 GB | 6+ GB | mods install flux2-klein-4b --variant gguf-q6-k |
| gguf-q5-k-m ⓘ | gguf | 3.1 GB | 6+ GB | mods install flux2-klein-4b --variant gguf-q5-k-m |
| gguf-q4-k-m ⓘ | gguf | 2.6 GB | 4+ GB | mods install flux2-klein-4b --variant gguf-q4-k-m |
| gguf-q3-k-m ⓘ | gguf | 2.1 GB | 4+ GB | mods install flux2-klein-4b --variant gguf-q3-k-m |
| gguf-q2-k ⓘ | gguf | 1.8 GB | 4+ GB | mods install flux2-klein-4b --variant gguf-q2-k |
Dependencies
These models are automatically installed when you run
mods install flux2-klein-4b. No extra steps needed — mods resolves and downloads all dependencies for you.