---
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-3.0-8b-instruct
model-index:
- name: granite-3.0-2b-instruct
  results:
  - task:
      type: text-generation
    dataset:
      name: IFEval
      type: instruction-following
    metrics:
    - type: pass@1
      value: 52.27
      name: pass@1
    - type: pass@1
      value: 8.22
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: AGI-Eval
      type: human-exams
    metrics:
    - type: pass@1
      value: 40.52
      name: pass@1
    - type: pass@1
      value: 65.82
      name: pass@1
    - type: pass@1
      value: 34.45
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: OBQA
      type: commonsense
    metrics:
    - type: pass@1
      value: 46.6
      name: pass@1
    - type: pass@1
      value: 71.21
      name: pass@1
    - type: pass@1
      value: 82.61
      name: pass@1
    - type: pass@1
      value: 77.51
      name: pass@1
    - type: pass@1
      value: 60.32
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: BoolQ
      type: reading-comprehension
    metrics:
    - type: pass@1
      value: 88.65
      name: pass@1
    - type: pass@1
      value: 21.58
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: ARC-C
      type: reasoning
    metrics:
    - type: pass@1
      value: 64.16
      name: pass@1
    - type: pass@1
      value: 33.81
      name: pass@1
    - type: pass@1
      value: 51.55
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: HumanEvalSynthesis
      type: code
    metrics:
    - type: pass@1
      value: 64.63
      name: pass@1
    - type: pass@1
      value: 57.16
      name: pass@1
    - type: pass@1
      value: 65.85
      name: pass@1
    - type: pass@1
      value: 49.6
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: GSM8K
      type: math
    metrics:
    - type: pass@1
      value: 68.99
      name: pass@1
    - type: pass@1
      value: 30.94
      name: pass@1
  - task:
      type: text-generation
    dataset:
      name: PAWS-X (7 langs)
      type: multilingual
    metrics:
    - type: pass@1
      value: 64.94
      name: pass@1
    - type: pass@1
      value: 48.2
      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-3.0-8b-instruct - GGUF

This repo contains GGUF format model files for [ibm-granite/granite-3.0-8b-instruct](https://huggingface.co/ibm-granite/granite-3.0-8b-instruct).

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


```
<|start_of_role|>system<|end_of_role|>{system_prompt}<|end_of_text|>
<|start_of_role|>user<|end_of_role|>{prompt}<|end_of_text|>
<|start_of_role|>assistant<|end_of_role|>
```

## Model file specification

| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [granite-3.0-8b-instruct-Q2_K.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q2_K.gguf) | Q2_K | 2.890 GB | smallest, significant quality loss - not recommended for most purposes |
| [granite-3.0-8b-instruct-Q3_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q3_K_S.gguf) | Q3_K_S | 3.346 GB | very small, high quality loss |
| [granite-3.0-8b-instruct-Q3_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q3_K_M.gguf) | Q3_K_M | 3.722 GB | very small, high quality loss |
| [granite-3.0-8b-instruct-Q3_K_L.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q3_K_L.gguf) | Q3_K_L | 4.051 GB | small, substantial quality loss |
| [granite-3.0-8b-instruct-Q4_0.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q4_0.gguf) | Q4_0 | 4.331 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [granite-3.0-8b-instruct-Q4_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q4_K_S.gguf) | Q4_K_S | 4.364 GB | small, greater quality loss |
| [granite-3.0-8b-instruct-Q4_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q4_K_M.gguf) | Q4_K_M | 4.603 GB | medium, balanced quality - recommended |
| [granite-3.0-8b-instruct-Q5_0.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q5_0.gguf) | Q5_0 | 5.259 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [granite-3.0-8b-instruct-Q5_K_S.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q5_K_S.gguf) | Q5_K_S | 5.259 GB | large, low quality loss - recommended |
| [granite-3.0-8b-instruct-Q5_K_M.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q5_K_M.gguf) | Q5_K_M | 5.399 GB | large, very low quality loss - recommended |
| [granite-3.0-8b-instruct-Q6_K.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q6_K.gguf) | Q6_K | 6.245 GB | very large, extremely low quality loss |
| [granite-3.0-8b-instruct-Q8_0.gguf](https://huggingface.co/tensorblock/granite-3.0-8b-instruct-GGUF/blob/main/granite-3.0-8b-instruct-Q8_0.gguf) | Q8_0 | 8.088 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-3.0-8b-instruct-GGUF --include "granite-3.0-8b-instruct-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-3.0-8b-instruct-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```