tensorblock/CodeMate-v0.1-GGUF

1---2language:3- en4license: llama25library_name: transformers6tags:7- CodeMate8- Code9- CodeLLaMa10- TensorBlock11- GGUF12pipeline_tag: text-generation13base_model: codemateai/CodeMate-v0.114model-index:15- name: CodeMate-v0.116  results:17  - task:18      type: text-generation19    dataset:20      name: HumanEval21      type: openai_humaneval22    metrics:23    - type: pass@124      value: 74.9%25      name: pass@126      verified: false27  - task:28      type: text-generation29      name: Text Generation30    dataset:31      name: AI2 Reasoning Challenge (25-Shot)32      type: ai2_arc33      config: ARC-Challenge34      split: test35      args:36        num_few_shot: 2537    metrics:38    - type: acc_norm39      value: 55.5540      name: normalized accuracy41    source:42      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.143      name: Open LLM Leaderboard44  - task:45      type: text-generation46      name: Text Generation47    dataset:48      name: HellaSwag (10-Shot)49      type: hellaswag50      split: validation51      args:52        num_few_shot: 1053    metrics:54    - type: acc_norm55      value: 78.0356      name: normalized accuracy57    source:58      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.159      name: Open LLM Leaderboard60  - task:61      type: text-generation62      name: Text Generation63    dataset:64      name: MMLU (5-Shot)65      type: cais/mmlu66      config: all67      split: test68      args:69        num_few_shot: 570    metrics:71    - type: acc72      value: 55.3173      name: accuracy74    source:75      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.176      name: Open LLM Leaderboard77  - task:78      type: text-generation79      name: Text Generation80    dataset:81      name: TruthfulQA (0-shot)82      type: truthful_qa83      config: multiple_choice84      split: validation85      args:86        num_few_shot: 087    metrics:88    - type: mc289      value: 48.6490    source:91      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.192      name: Open LLM Leaderboard93  - task:94      type: text-generation95      name: Text Generation96    dataset:97      name: Winogrande (5-shot)98      type: winogrande99      config: winogrande_xl100      split: validation101      args:102        num_few_shot: 5103    metrics:104    - type: acc105      value: 72.61106      name: accuracy107    source:108      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.1109      name: Open LLM Leaderboard110  - task:111      type: text-generation112      name: Text Generation113    dataset:114      name: GSM8k (5-shot)115      type: gsm8k116      config: main117      split: test118      args:119        num_few_shot: 5120    metrics:121    - type: acc122      value: 40.18123      name: accuracy124    source:125      url: https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderboard?query=codemateai/CodeMate-v0.1126      name: Open LLM Leaderboard127---128 129<div style="width: auto; margin-left: auto; margin-right: auto">130<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">131</div>132 133[![Website](https://img.shields.io/badge/Website-tensorblock.co-blue?logo=google-chrome&logoColor=white)](https://tensorblock.co)134[![Twitter](https://img.shields.io/twitter/follow/tensorblock_aoi?style=social)](https://twitter.com/tensorblock_aoi)135[![Discord](https://img.shields.io/badge/Discord-Join%20Us-5865F2?logo=discord&logoColor=white)](https://discord.gg/Ej5NmeHFf2)136[![GitHub](https://img.shields.io/badge/GitHub-TensorBlock-black?logo=github&logoColor=white)](https://github.com/TensorBlock)137[![Telegram](https://img.shields.io/badge/Telegram-Group-blue?logo=telegram)](https://t.me/TensorBlock)138 139 140## codemateai/CodeMate-v0.1 - GGUF141 142This repo contains GGUF format model files for [codemateai/CodeMate-v0.1](https://huggingface.co/codemateai/CodeMate-v0.1).143 144The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b4242](https://github.com/ggerganov/llama.cpp/commit/a6744e43e80f4be6398fc7733a01642c846dce1d).145 146## Our projects147<table border="1" cellspacing="0" cellpadding="10">148  <tr>149    <th colspan="2" style="font-size: 25px;">Forge</th>150  </tr>151  <tr>152    <th colspan="2">153      <img src="https://imgur.com/faI5UKh.jpeg" alt="Forge Project" width="900"/>154    </th>155  </tr>156  <tr>157    <th colspan="2">An OpenAI-compatible multi-provider routing layer.</th>158  </tr>159  <tr>160    <th colspan="2">161      <a href="https://github.com/TensorBlock/forge" target="_blank" style="162        display: inline-block;163        padding: 8px 16px;164        background-color: #FF7F50;165        color: white;166        text-decoration: none;167        border-radius: 6px;168        font-weight: bold;169        font-family: sans-serif;170      ">🚀 Try it now! 🚀</a>171    </th>172  </tr>173 174  <tr>175    <th style="font-size: 25px;">Awesome MCP Servers</th>176    <th style="font-size: 25px;">TensorBlock Studio</th>177  </tr>178  <tr>179    <th><img src="https://imgur.com/2Xov7B7.jpeg" alt="MCP Servers" width="450"/></th>180    <th><img src="https://imgur.com/pJcmF5u.jpeg" alt="Studio" width="450"/></th>181  </tr>182  <tr>183    <th>A comprehensive collection of Model Context Protocol (MCP) servers.</th>184    <th>A lightweight, open, and extensible multi-LLM interaction studio.</th>185  </tr>186  <tr>187    <th>188      <a href="https://github.com/TensorBlock/awesome-mcp-servers" target="_blank" style="189        display: inline-block;190        padding: 8px 16px;191        background-color: #FF7F50;192        color: white;193        text-decoration: none;194        border-radius: 6px;195        font-weight: bold;196        font-family: sans-serif;197      ">👀 See what we built 👀</a>198    </th>199    <th>200      <a href="https://github.com/TensorBlock/TensorBlock-Studio" target="_blank" style="201        display: inline-block;202        padding: 8px 16px;203        background-color: #FF7F50;204        color: white;205        text-decoration: none;206        border-radius: 6px;207        font-weight: bold;208        font-family: sans-serif;209      ">👀 See what we built 👀</a>210    </th>211  </tr>212</table>213## Prompt template214 215```216 217```218 219## Model file specification220 221| Filename | Quant type | File Size | Description |222| -------- | ---------- | --------- | ----------- |223| [CodeMate-v0.1-Q2_K.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q2_K.gguf) | Q2_K | 12.506 GB | smallest, significant quality loss - not recommended for most purposes |224| [CodeMate-v0.1-Q3_K_S.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q3_K_S.gguf) | Q3_K_S | 14.605 GB | very small, high quality loss |225| [CodeMate-v0.1-Q3_K_M.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q3_K_M.gguf) | Q3_K_M | 16.306 GB | very small, high quality loss |226| [CodeMate-v0.1-Q3_K_L.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q3_K_L.gguf) | Q3_K_L | 17.772 GB | small, substantial quality loss |227| [CodeMate-v0.1-Q4_0.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q4_0.gguf) | Q4_0 | 19.052 GB | legacy; small, very high quality loss - prefer using Q3_K_M |228| [CodeMate-v0.1-Q4_K_S.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q4_K_S.gguf) | Q4_K_S | 19.192 GB | small, greater quality loss |229| [CodeMate-v0.1-Q4_K_M.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q4_K_M.gguf) | Q4_K_M | 20.220 GB | medium, balanced quality - recommended |230| [CodeMate-v0.1-Q5_0.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q5_0.gguf) | Q5_0 | 23.237 GB | legacy; medium, balanced quality - prefer using Q4_K_M |231| [CodeMate-v0.1-Q5_K_S.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q5_K_S.gguf) | Q5_K_S | 23.237 GB | large, low quality loss - recommended |232| [CodeMate-v0.1-Q5_K_M.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q5_K_M.gguf) | Q5_K_M | 23.839 GB | large, very low quality loss - recommended |233| [CodeMate-v0.1-Q6_K.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q6_K.gguf) | Q6_K | 27.684 GB | very large, extremely low quality loss |234| [CodeMate-v0.1-Q8_0.gguf](https://huggingface.co/tensorblock/CodeMate-v0.1-GGUF/blob/main/CodeMate-v0.1-Q8_0.gguf) | Q8_0 | 35.856 GB | very large, extremely low quality loss - not recommended |235 236 237## Downloading instruction238 239### Command line240 241Firstly, install Huggingface Client242 243```shell244pip install -U "huggingface_hub[cli]"245```246 247Then, downoad the individual model file the a local directory248 249```shell250huggingface-cli download tensorblock/CodeMate-v0.1-GGUF --include "CodeMate-v0.1-Q2_K.gguf" --local-dir MY_LOCAL_DIR251```252 253If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:254 255```shell256huggingface-cli download tensorblock/CodeMate-v0.1-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'257```258