Update README.md
Browse files
README.md
CHANGED
|
@@ -32,6 +32,13 @@ tags:
|
|
| 32 |
- example workflow for [gguf](https://github.com/calcuis/comfy/blob/main/workflow-sd3.5m-gguf.json) (if it doesn't work, upgrade your pack: ggc y) 👻
|
| 33 |
- example workflow for the original [safetensors](https://github.com/calcuis/comfy/blob/main/workflow-sd3.5m-safetensors.json) 🎃
|
| 34 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 35 |
### Upper tier sets
|
| 36 |
- sd3.5-[large](https://huggingface.co/calcuis/sd3.5-large-gguf) (recommended)
|
| 37 |
- sd3.5-[turbo](https://huggingface.co/calcuis/sd3.5-large-turbo)
|
|
|
|
| 32 |
- example workflow for [gguf](https://github.com/calcuis/comfy/blob/main/workflow-sd3.5m-gguf.json) (if it doesn't work, upgrade your pack: ggc y) 👻
|
| 33 |
- example workflow for the original [safetensors](https://github.com/calcuis/comfy/blob/main/workflow-sd3.5m-safetensors.json) 🎃
|
| 34 |
|
| 35 |
+
### Bug reports (or brief review)
|
| 36 |
+
- t/q1_0 and t/q2_0; doesn't work; encountering np.uint32(34 & 35) is not a valid GGMLQuantizationType error
|
| 37 |
+
- q2_k is super fast; but picture looks like x-ray film; might be good for medical research or abstract stylist/artist
|
| 38 |
+
- q3 family is very fast; finger problem can be easily detected but picture quality is interestingly good
|
| 39 |
+
- notice the same file size in 0 and s models; still keep them all here; see who can help out to look into details
|
| 40 |
+
- overall, q4 and above should be no problem for general to high quality production
|
| 41 |
+
|
| 42 |
### Upper tier sets
|
| 43 |
- sd3.5-[large](https://huggingface.co/calcuis/sd3.5-large-gguf) (recommended)
|
| 44 |
- sd3.5-[turbo](https://huggingface.co/calcuis/sd3.5-large-turbo)
|