Krea2_turbo_diff_loras
These LoRAs were extracted from fine-tuned checkpoints. I did this primarily for myself because LoRAs are easier to work with: they can be mixed in real time and assigned different weights/strengths. This also saves space on your local computer. They work well at 100–150% strength. Choose a rank according to your taste and your hardware capabilities (the lower the rank, the less memory you need). However, some of my...
Params
12.8 B
Context
—
Downloads 30d
92 K
Likes
83
Download history
daily snapshots · 84 days
▲ 39 K in the last 30 days (74.5%)
92 K56 K
Oct 12Oct 22Nov 1Nov 10
92 K18 K
Aug 19Sep 16Oct 14Nov 10
Can you run it?
Estimated VRAM at 8K context unless noted. Pick your hardware to see the verdict per quantization.
| File | Quant | Size | Est. VRAM | Verdict on RTX 4090 · 24 GB |
|---|---|---|---|---|
| model.safetensors | fp16 | 786.4 GB | 867.4 GB | ❌ Won’t fit |
Estimate: file size × 1.1 + KV cache at 8K + 0.5 GB overhead. Not a benchmark — how we calculate this.
Specifications
- Parameters
- 12.8 B
- First seen on the Hub
- 2026-08-01
- Base model
- Krea-2-Turbo
- Training datasets
- undisclosed
- Added to our catalog
- 2026-08-19
Compare with any text-to-image model