Update README.md
Browse files
README.md
CHANGED
|
@@ -1,5 +1,28 @@
|
|
| 1 |
-
---
|
| 2 |
-
|
| 3 |
-
|
| 4 |
-
|
| 5 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: black-forest-labs/FLUX.1-dev
|
| 3 |
+
library_name: gguf
|
| 4 |
+
license: other
|
| 5 |
+
license_name: flux-1-dev-non-commercial-license
|
| 6 |
+
license_link: LICENSE.md
|
| 7 |
+
quantized_by: city96
|
| 8 |
+
tags:
|
| 9 |
+
- text-to-image
|
| 10 |
+
- image-generation
|
| 11 |
+
- flux
|
| 12 |
+
---
|
| 13 |
+
|
| 14 |
+
Flux.1-dev in **bfloat16**, exactly the same tensors as in black-forest-labs' original safetensors.
|
| 15 |
+
AFAIK no inference software supports bf16 FLUX yet, but maybe this file can help.
|
| 16 |
+
I have no clue if it would improve anything, but bf16 –> fp16 conversion is lossy, so... maybe?
|
| 17 |
+
|
| 18 |
+
This is a direct GGUF conversion of [black-forest-labs/FLUX.1-dev](https://huggingface.co/black-forest-labs/FLUX.1-dev/tree/main)
|
| 19 |
+
|
| 20 |
+
As this is a quantized model not a finetune, all the same restrictions/original license terms still apply.
|
| 21 |
+
|
| 22 |
+
The model files can be used with the [ComfyUI-GGUF](https://github.com/city96/ComfyUI-GGUF) custom node.
|
| 23 |
+
|
| 24 |
+
Place model files in `ComfyUI/models/unet` - see the GitHub readme for further install instructions.
|
| 25 |
+
|
| 26 |
+
Please refer to [this chart](https://github.com/ggerganov/llama.cpp/blob/master/examples/perplexity/README.md#llama-3-8b-scoreboard) for a basic overview of quantization types.
|
| 27 |
+
|
| 28 |
+
(Model card mostly copied from [city96/FLUX.1-dev-gguf](https://huggingface.co/city96/FLUX.1-dev-gguf) - which contains conventional and useful GGUF files.)
|