upstream-archive/stable-diffusion.cpp
1
0
Fork 0
mirror of https://github.com/leejet/stable-diffusion.cpp.git synced 2026-08-02 16:20:41 -05:00
Code Issues Packages Projects Releases Wiki Activity

fix: clarify lora quant support and small fixes (#792)

Browse Source
This commit is contained in:
Markus Hartung
2025-09-08 16:39:25 +02:00
committed by GitHub
parent c648001030
commit abb115cd02
5 changed files with 49 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

28
docs/lora.md
View File

@@ -10,4 +10,30 @@ Here's a simple example:
./bin/sd -m ../models/v1-5-pruned-emaonly.safetensors -p "a lovely cat<lora:marblesh:1>" --lora-model-dir ../models
```
`../models/marblesh.safetensors` or `../models/marblesh.ckpt` will be applied to the model
`../models/marblesh.safetensors` or `../models/marblesh.ckpt` will be applied to the model
# Support matrix
> ℹ️ CUDA `get_rows` support is defined here:
> [ggml-org/ggml/src/ggml-cuda/getrows.cu#L156](https://github.com/ggml-org/ggml/blob/7dee1d6a1e7611f238d09be96738388da97c88ed/src/ggml-cuda/getrows.cu#L156)
> Currently only the basic types + Q4/Q5/Q8 are implemented. K-quants are **not** supported.
NOTE: The other backends may have different support.
| Quant / Type | CUDA |
|--------------|------|
| F32 | ✔️ |
| F16 | ✔️ |
| BF16 | ✔️ |
| I32 | ✔️ |
| Q4_0 | ✔️ |
| Q4_1 | ✔️ |
| Q5_0 | ✔️ |
| Q5_1 | ✔️ |
| Q8_0 | ✔️ |
| Q2_K | ❌ |
| Q3_K | ❌ |
| Q4_K | ❌ |
| Q5_K | ❌ |
| Q6_K | ❌ |
| Q8_K | ❌ |