Upload README.md with huggingface_hub
Browse files
README.md
ADDED
|
@@ -0,0 +1,46 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
---
|
| 2 |
+
base_model: google/gemma-3-270m-it
|
| 3 |
+
base_model_relation: quantized
|
| 4 |
+
quantized_by: ArtusDev
|
| 5 |
+
---
|
| 6 |
+
|
| 7 |
+
## EXL3 Quants of google/gemma-3-270m-it
|
| 8 |
+
|
| 9 |
+
EXL3 quants of [google/gemma-3-270m-it](https://huggingface.co/google/gemma-3-270m-it) using <a href="https://github.com/turboderp-org/exllamav3/">exllamav3</a> for quantization.
|
| 10 |
+
|
| 11 |
+
### Quants
|
| 12 |
+
| Quant(Revision) | Bits per Weight | Head Bits |
|
| 13 |
+
| -------- | ---------- | --------- |
|
| 14 |
+
| [1.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/1.0bpw_H8) | 1.0 | 8 |
|
| 15 |
+
| [1.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/1.5bpw_H8) | 1.5 | 8 |
|
| 16 |
+
| [2.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/2.0bpw_H8) | 2.0 | 8 |
|
| 17 |
+
| [2.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/2.5bpw_H8) | 2.5 | 8 |
|
| 18 |
+
| [3.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/3.0bpw_H8) | 3.0 | 8 |
|
| 19 |
+
| [3.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/3.5bpw_H8) | 3.5 | 8 |
|
| 20 |
+
| [4.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/4.0bpw_H8) | 4.0 | 8 |
|
| 21 |
+
| [4.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/4.5bpw_H8) | 4.5 | 8 |
|
| 22 |
+
| [5.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/5.0bpw_H8) | 5.0 | 8 |
|
| 23 |
+
| [5.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/5.5bpw_H8) | 5.5 | 8 |
|
| 24 |
+
| [6.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/6.0bpw_H8) | 6.0 | 8 |
|
| 25 |
+
| [6.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/6.5bpw_H8) | 6.5 | 8 |
|
| 26 |
+
| [7.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/7.0bpw_H8) | 7.0 | 8 |
|
| 27 |
+
| [7.5_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/7.5bpw_H8) | 7.5 | 8 |
|
| 28 |
+
| [8.0_H8](https://huggingface.co/ArtusDev/google_gemma-3-270m-it-EXL3/tree/8.0bpw_H8) | 8.0 | 8 |
|
| 29 |
+
|
| 30 |
+
### Downloading quants with huggingface-cli
|
| 31 |
+
|
| 32 |
+
<details>
|
| 33 |
+
<summary>Click to view download instructions</summary>
|
| 34 |
+
|
| 35 |
+
Install hugginface-cli:
|
| 36 |
+
|
| 37 |
+
```bash
|
| 38 |
+
pip install -U "huggingface_hub[cli]"
|
| 39 |
+
```
|
| 40 |
+
|
| 41 |
+
Download quant by targeting the specific quant revision (branch):
|
| 42 |
+
|
| 43 |
+
```
|
| 44 |
+
huggingface-cli download ArtusDev/google_gemma-3-270m-it-EXL3 --revision "5.0bpw_H6" --local-dir ./
|
| 45 |
+
```
|
| 46 |
+
</details>
|