tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF
076
1---2library_name: transformers3license: other4license_name: hyperclovax-seed5license_link: LICENSE6datasets:7- exp-models/Open-Reasoner-Zero-orz-math-57k-collected-Korean8language:9- ko10base_model: werty1248/HyperCLOVAX-1.5B-Reasoning-RFT11tags:12- TensorBlock13- GGUF14---15 16<div style="width: auto; margin-left: auto; margin-right: auto">17<img src="https://i.imgur.com/jC7kdl8.jpeg" alt="TensorBlock" style="width: 100%; min-width: 400px; display: block; margin: auto;">18</div>19 20[](https://tensorblock.co)21[](https://twitter.com/tensorblock_aoi)22[](https://discord.gg/Ej5NmeHFf2)23[](https://github.com/TensorBlock)24[](https://t.me/TensorBlock)25 26 27## werty1248/HyperCLOVAX-1.5B-Reasoning-RFT - GGUF28 29<div style="text-align: left; margin: 20px 0;">30 <a href="https://discord.com/invite/Ej5NmeHFf2" style="display: inline-block; padding: 10px 20px; background-color: #5865F2; color: white; text-decoration: none; border-radius: 5px; font-weight: bold;">31 Join our Discord to learn more about what we're building ↗32 </a>33</div>34 35This repo contains GGUF format model files for [werty1248/HyperCLOVAX-1.5B-Reasoning-RFT](https://huggingface.co/werty1248/HyperCLOVAX-1.5B-Reasoning-RFT).36 37The files were quantized using machines provided by [TensorBlock](https://tensorblock.co/), and they are compatible with llama.cpp as of [commit b5753](https://github.com/ggml-org/llama.cpp/commit/73e53dc834c0a2336cd104473af6897197b96277).38 39## Our projects40<table border="1" cellspacing="0" cellpadding="10">41 <tr>42 <th colspan="2" style="font-size: 25px;">Forge</th>43 </tr>44 <tr>45 <th colspan="2">46 <img src="https://imgur.com/faI5UKh.jpeg" alt="Forge Project" width="900"/>47 </th>48 </tr>49 <tr>50 <th colspan="2">An OpenAI-compatible multi-provider routing layer.</th>51 </tr>52 <tr>53 <th colspan="2">54 <a href="https://github.com/TensorBlock/forge" target="_blank" style="55 display: inline-block;56 padding: 8px 16px;57 background-color: #FF7F50;58 color: white;59 text-decoration: none;60 border-radius: 6px;61 font-weight: bold;62 font-family: sans-serif;63 ">🚀 Try it now! 🚀</a>64 </th>65 </tr>66 67 <tr>68 <th style="font-size: 25px;">Awesome MCP Servers</th>69 <th style="font-size: 25px;">TensorBlock Studio</th>70 </tr>71 <tr>72 <th><img src="https://imgur.com/2Xov7B7.jpeg" alt="MCP Servers" width="450"/></th>73 <th><img src="https://imgur.com/pJcmF5u.jpeg" alt="Studio" width="450"/></th>74 </tr>75 <tr>76 <th>A comprehensive collection of Model Context Protocol (MCP) servers.</th>77 <th>A lightweight, open, and extensible multi-LLM interaction studio.</th>78 </tr>79 <tr>80 <th>81 <a href="https://github.com/TensorBlock/awesome-mcp-servers" target="_blank" style="82 display: inline-block;83 padding: 8px 16px;84 background-color: #FF7F50;85 color: white;86 text-decoration: none;87 border-radius: 6px;88 font-weight: bold;89 font-family: sans-serif;90 ">👀 See what we built 👀</a>91 </th>92 <th>93 <a href="https://github.com/TensorBlock/TensorBlock-Studio" target="_blank" style="94 display: inline-block;95 padding: 8px 16px;96 background-color: #FF7F50;97 color: white;98 text-decoration: none;99 border-radius: 6px;100 font-weight: bold;101 font-family: sans-serif;102 ">👀 See what we built 👀</a>103 </th>104 </tr>105</table>106 107## Prompt template108 109```110<|im_start|>system111{system_prompt}<|im_end|>112<|im_start|>user113{prompt}<|im_end|>114<|im_start|>assistant115```116 117## Model file specification118 119| Filename | Quant type | File Size | Description |120| -------- | ---------- | --------- | ----------- |121| [HyperCLOVAX-1.5B-Reasoning-RFT-Q2_K.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q2_K.gguf) | Q2_K | 0.687 GB | smallest, significant quality loss - not recommended for most purposes |122| [HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_S.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_S.gguf) | Q3_K_S | 0.774 GB | very small, high quality loss |123| [HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_M.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_M.gguf) | Q3_K_M | 0.843 GB | very small, high quality loss |124| [HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_L.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q3_K_L.gguf) | Q3_K_L | 0.904 GB | small, substantial quality loss |125| [HyperCLOVAX-1.5B-Reasoning-RFT-Q4_0.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q4_0.gguf) | Q4_0 | 0.955 GB | legacy; small, very high quality loss - prefer using Q3_K_M |126| [HyperCLOVAX-1.5B-Reasoning-RFT-Q4_K_S.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q4_K_S.gguf) | Q4_K_S | 0.961 GB | small, greater quality loss |127| [HyperCLOVAX-1.5B-Reasoning-RFT-Q4_K_M.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q4_K_M.gguf) | Q4_K_M | 1.007 GB | medium, balanced quality - recommended |128| [HyperCLOVAX-1.5B-Reasoning-RFT-Q5_0.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q5_0.gguf) | Q5_0 | 1.125 GB | legacy; medium, balanced quality - prefer using Q4_K_M |129| [HyperCLOVAX-1.5B-Reasoning-RFT-Q5_K_S.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q5_K_S.gguf) | Q5_K_S | 1.125 GB | large, low quality loss - recommended |130| [HyperCLOVAX-1.5B-Reasoning-RFT-Q5_K_M.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q5_K_M.gguf) | Q5_K_M | 1.151 GB | large, very low quality loss - recommended |131| [HyperCLOVAX-1.5B-Reasoning-RFT-Q6_K.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q6_K.gguf) | Q6_K | 1.305 GB | very large, extremely low quality loss |132| [HyperCLOVAX-1.5B-Reasoning-RFT-Q8_0.gguf](https://huggingface.co/tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF/blob/main/HyperCLOVAX-1.5B-Reasoning-RFT-Q8_0.gguf) | Q8_0 | 1.689 GB | very large, extremely low quality loss - not recommended |133 134 135## Downloading instruction136 137### Command line138 139Firstly, install Huggingface Client140 141```shell142pip install -U "huggingface_hub[cli]"143```144 145Then, downoad the individual model file the a local directory146 147```shell148huggingface-cli download tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF --include "HyperCLOVAX-1.5B-Reasoning-RFT-Q2_K.gguf" --local-dir MY_LOCAL_DIR149```150 151If you wanna download multiple model files with a pattern (e.g., `*Q4_K*gguf`), you can try:152 153```shell154huggingface-cli download tensorblock/werty1248_HyperCLOVAX-1.5B-Reasoning-RFT-GGUF --local-dir MY_LOCAL_DIR --local-dir-use-symlinks False --include='*Q4_K*gguf'155```156