Update README.md
Browse files
README.md
CHANGED
@@ -1,6 +1,6 @@
|
|
1 |
---
|
2 |
quantized_by: bartowski
|
3 |
-
pipeline_tag: text-
|
4 |
license: gemma
|
5 |
base_model: mlabonne/gemma-3-27b-it-abliterated
|
6 |
---
|
@@ -25,14 +25,15 @@ Run them directly with [llama.cpp](https://github.com/ggerganov/llama.cpp), or a
|
|
25 |
|
26 |
{prompt}<end_of_turn>
|
27 |
<start_of_turn>model
|
28 |
-
|
29 |
-
<start_of_turn>model
|
30 |
```
|
31 |
|
32 |
## Download a file (not the whole branch) from below:
|
33 |
|
34 |
| Filename | Quant type | File Size | Split | Description |
|
35 |
| -------- | ---------- | --------- | ----- | ----------- |
|
|
|
|
|
36 |
| [gemma-3-27b-it-abliterated-bf16.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/tree/main/mlabonne_gemma-3-27b-it-abliterated-bf16) | bf16 | 54.03GB | true | Full BF16 weights. |
|
37 |
| [gemma-3-27b-it-abliterated-Q8_0.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/blob/main/mlabonne_gemma-3-27b-it-abliterated-Q8_0.gguf) | Q8_0 | 28.71GB | false | Extremely high quality, generally unneeded but max available quant. |
|
38 |
| [gemma-3-27b-it-abliterated-Q6_K_L.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/blob/main/mlabonne_gemma-3-27b-it-abliterated-Q6_K_L.gguf) | Q6_K_L | 22.51GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
|
@@ -173,4 +174,4 @@ Thank you ZeroWw for the inspiration to experiment with embed/output.
|
|
173 |
|
174 |
Thank you to LM Studio for sponsoring my work.
|
175 |
|
176 |
-
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
|
|
|
1 |
---
|
2 |
quantized_by: bartowski
|
3 |
+
pipeline_tag: image-text-to-text
|
4 |
license: gemma
|
5 |
base_model: mlabonne/gemma-3-27b-it-abliterated
|
6 |
---
|
|
|
25 |
|
26 |
{prompt}<end_of_turn>
|
27 |
<start_of_turn>model
|
28 |
+
|
|
|
29 |
```
|
30 |
|
31 |
## Download a file (not the whole branch) from below:
|
32 |
|
33 |
| Filename | Quant type | File Size | Split | Description |
|
34 |
| -------- | ---------- | --------- | ----- | ----------- |
|
35 |
+
| [mmproj-gemma-3-27b-it-abliterated-f32.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-GGUF/blob/main/mmproj-mlabonne_gemma-3-27b-it-f32.gguf) | f32 | 1.69GB | false | F32 format MMPROJ file, required for vision. |
|
36 |
+
| [mmproj-gemma-3-27b-it-abliterated-f16.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-GGUF/blob/main/mmproj-mlabonne_gemma-3-27b-it-f16.gguf) | f16 | 858MB | false | F16 format MMPROJ file, required for vision. |
|
37 |
| [gemma-3-27b-it-abliterated-bf16.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/tree/main/mlabonne_gemma-3-27b-it-abliterated-bf16) | bf16 | 54.03GB | true | Full BF16 weights. |
|
38 |
| [gemma-3-27b-it-abliterated-Q8_0.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/blob/main/mlabonne_gemma-3-27b-it-abliterated-Q8_0.gguf) | Q8_0 | 28.71GB | false | Extremely high quality, generally unneeded but max available quant. |
|
39 |
| [gemma-3-27b-it-abliterated-Q6_K_L.gguf](https://huggingface.co/bartowski/mlabonne_gemma-3-27b-it-abliterated-GGUF/blob/main/mlabonne_gemma-3-27b-it-abliterated-Q6_K_L.gguf) | Q6_K_L | 22.51GB | false | Uses Q8_0 for embed and output weights. Very high quality, near perfect, *recommended*. |
|
|
|
174 |
|
175 |
Thank you to LM Studio for sponsoring my work.
|
176 |
|
177 |
+
Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski
|