---
language:
- en
license: cc-by-nc-sa-4.0
datasets:
- Intel/orca_dpo_pairs
- argilla/distilabel-math-preference-dpo
- kyujinpy/orca_math_dpo
pipeline_tag: text-generation
base_model: kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
tags:
- TensorBlock
- GGUF
model-index:
- name: Sakura-SOLRCA-Math-Instruct-DPO-v1
  results:
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: AI2 Reasoning Challenge (25-Shot)
      type: ai2_arc
      config: ARC-Challenge
      split: test
      args:
        num_few_shot: 25
    metrics:
    - type: acc_norm
      value: 71.25
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: HellaSwag (10-Shot)
      type: hellaswag
      split: validation
      args:
        num_few_shot: 10
    metrics:
    - type: acc_norm
      value: 88.48
      name: normalized accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: MMLU (5-Shot)
      type: cais/mmlu
      config: all
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 66.21
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: TruthfulQA (0-shot)
      type: truthful_qa
      config: multiple_choice
      split: validation
      args:
        num_few_shot: 0
    metrics:
    - type: mc2
      value: 72.12
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: Winogrande (5-shot)
      type: winogrande
      config: winogrande_xl
      split: validation
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 82.87
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
  - task:
      type: text-generation
      name: Text Generation
    dataset:
      name: GSM8k (5-shot)
      type: gsm8k
      config: main
      split: test
      args:
        num_few_shot: 5
    metrics:
    - type: acc
      value: 63.84
      name: accuracy
    source:
      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1
      name: Open LLM Leaderboard
---

<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>

## kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1 - GGUF

This repo contains GGUF format model files for [kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1](https://huggingface.co/kyujinpy/Sakura-SOLRCA-Math-Instruct-DPO-v1).

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).

## Prompt template

```
### System:
{system_prompt}

### User:
{prompt}

### Assistant:
```

## Model file specification

| Filename | Quant type | File Size | Description |
| -------- | ---------- | --------- | ----------- |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q2_K.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q2_K.gguf) | Q2_K | 3.728 GB | smallest, significant quality loss - not recommended for most purposes |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_S.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_S.gguf) | Q3_K_S | 4.344 GB | very small, high quality loss |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_M.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_M.gguf) | Q3_K_M | 4.839 GB | very small, high quality loss |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_L.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q3_K_L.gguf) | Q3_K_L | 5.263 GB | small, substantial quality loss |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_0.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_0.gguf) | Q4_0 | 5.655 GB | legacy; small, very high quality loss - prefer using Q3_K_M |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_K_S.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_K_S.gguf) | Q4_K_S | 5.698 GB | small, greater quality loss |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_K_M.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q4_K_M.gguf) | Q4_K_M | 6.018 GB | medium, balanced quality - recommended |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_0.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_0.gguf) | Q5_0 | 6.889 GB | legacy; medium, balanced quality - prefer using Q4_K_M |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_K_S.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_K_S.gguf) | Q5_K_S | 6.889 GB | large, low quality loss - recommended |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_K_M.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q5_K_M.gguf) | Q5_K_M | 7.076 GB | large, very low quality loss - recommended |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q6_K.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q6_K.gguf) | Q6_K | 8.200 GB | very large, extremely low quality loss |
| [Sakura-SOLRCA-Math-Instruct-DPO-v1-Q8_0.gguf](https://huggingface.co/tensorblock/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF/tree/main/Sakura-SOLRCA-Math-Instruct-DPO-v1-Q8_0.gguf) | Q8_0 | 10.621 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/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF --include "Sakura-SOLRCA-Math-Instruct-DPO-v1-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/Sakura-SOLRCA-Math-Instruct-DPO-v1-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'
```