|
--- |
|
pipeline_tag: text-generation |
|
inference: false |
|
license: apache-2.0 |
|
library_name: transformers |
|
tags: |
|
- language |
|
- granite-3.0 |
|
- TensorBlock |
|
- GGUF |
|
base_model: ibm-granite/granite-3.0-2b-base |
|
model-index: |
|
- name: granite-3.0-2b-base |
|
results: |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: MMLU |
|
type: human-exams |
|
metrics: |
|
- type: pass@1 |
|
value: 55.0 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 23.79 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 22.56 |
|
name: pass@1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: WinoGrande |
|
type: commonsense |
|
metrics: |
|
- type: pass@1 |
|
value: 74.9 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 43.0 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 59.84 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 79.27 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 77.65 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 39.9 |
|
name: pass@1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: BoolQ |
|
type: reading-comprehension |
|
metrics: |
|
- type: pass@1 |
|
value: 81.35 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 25.22 |
|
name: pass@1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: ARC-C |
|
type: reasoning |
|
metrics: |
|
- type: pass@1 |
|
value: 54.27 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 30.58 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 40.69 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 34.34 |
|
name: pass@1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: HumanEval |
|
type: code |
|
metrics: |
|
- type: pass@1 |
|
value: 38.41 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 35.4 |
|
name: pass@1 |
|
- task: |
|
type: text-generation |
|
dataset: |
|
name: GSM8K |
|
type: math |
|
metrics: |
|
- type: pass@1 |
|
value: 47.23 |
|
name: pass@1 |
|
- type: pass@1 |
|
value: 19.46 |
|
name: pass@1 |
|
--- |
|
|
|
<div style="width: auto; margin-left: auto; margin-right: auto"> |
|
<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;"> |
|
</div> |
|
<div style="display: flex; justify-content: space-between; width: 100%;"> |
|
<div style="display: flex; flex-direction: column; align-items: flex-start;"> |
|
<p style="margin-top: 0.5em; margin-bottom: 0em;"> |
|
Feedback and support: TensorBlock's <a href="https://x.com/tensorblock_aoi">Twitter/X</a>, <a href="https://t.me/TensorBlock">Telegram Group</a> and <a href="https://x.com/tensorblock_aoi">Discord server</a> |
|
</p> |
|
</div> |
|
</div> |
|
|
|
## ibm-granite/granite-3.0-2b-base - GGUF |
|
|
|
This repo contains GGUF format model files for [ibm-granite/granite-3.0-2b-base](https://huggingface.co/ibm-granite/granite-3.0-2b-base). |
|
|
|
The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4011](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d). |
|
|
|
<div style="text-align: left; margin: 20px 0;"> |
|
<a href="https://tensorblock.co/waitlist/client" style="display: inline-block; padding: 10px 20px; background-color: #007bff; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;"> |
|
Run them on the TensorBlock client using your local machine ↗ |
|
</a> |
|
</div> |
|
|
|
## Prompt template |
|
|
|
``` |
|
|
|
``` |
|
|
|
## Model file specification |
|
|
|
| Filename | Quant type | File Size | Description | |
|
| -------- | ---------- | --------- | ----------- | |
|
| [granite-3.0-2b-base-Q2_K.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q2_K.gguf) | Q2_K | 0.978 GB | smallest, significant quality loss - not recommended for most purposes | |
|
| [granite-3.0-2b-base-Q3_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q3_K_S.gguf) | Q3_K_S | 1.130 GB | very small, high quality loss | |
|
| [granite-3.0-2b-base-Q3_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q3_K_M.gguf) | Q3_K_M | 1.252 GB | very small, high quality loss | |
|
| [granite-3.0-2b-base-Q3_K_L.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q3_K_L.gguf) | Q3_K_L | 1.357 GB | small, substantial quality loss | |
|
| [granite-3.0-2b-base-Q4_0.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q4_0.gguf) | Q4_0 | 1.453 GB | legacy; small, very high quality loss - prefer using Q3_K_M | |
|
| [granite-3.0-2b-base-Q4_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q4_K_S.gguf) | Q4_K_S | 1.464 GB | small, greater quality loss | |
|
| [granite-3.0-2b-base-Q4_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q4_K_M.gguf) | Q4_K_M | 1.545 GB | medium, balanced quality - recommended | |
|
| [granite-3.0-2b-base-Q5_0.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q5_0.gguf) | Q5_0 | 1.757 GB | legacy; medium, balanced quality - prefer using Q4_K_M | |
|
| [granite-3.0-2b-base-Q5_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q5_K_S.gguf) | Q5_K_S | 1.757 GB | large, low quality loss - recommended | |
|
| [granite-3.0-2b-base-Q5_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q5_K_M.gguf) | Q5_K_M | 1.805 GB | large, very low quality loss - recommended | |
|
| [granite-3.0-2b-base-Q6_K.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q6_K.gguf) | Q6_K | 2.081 GB | very large, extremely low quality loss | |
|
| [granite-3.0-2b-base-Q8_0.gguf](https://huggingface.co/tensorblock/granite-3.0-2b-base-GGUF/blob/main/granite-3.0-2b-base-Q8_0.gguf) | Q8_0 | 2.694 GB | very large, extremely low quality loss - not recommended | |
|
|
|
|
|
## Downloading instruction |
|
|
|
### Command line |
|
|
|
Firstly, install Huggingface Client |
|
|
|
```shell |
|
pip install -U "huggingface_hub[cli]" |
|
``` |
|
|
|
Then, downoad the individual model file the a local directory |
|
|
|
```shell |
|
huggingface-cli download tensorblock/granite-3.0-2b-base-GGUF --include "granite-3.0-2b-base-Q2_K.gguf" --local-dir MY_LOCAL_DIR |
|
``` |
|
|
|
If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try: |
|
|
|
```shell |
|
huggingface-cli download tensorblock/granite-3.0-2b-base-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf' |
|
``` |
|
|
