lewdiculous/mn-12b-lyra-v4-gguf-iq-imatrix IQ4_XS GGUF - Free GGUF Download is indexed on GraySoft with repository links, GGUF quant files, and Hugging Face metadata. This page helps you pick a local model for guIDE or other runtimes. See related models in the same shard below.
lewdiculous/mn-12b-lyra-v4-gguf-iq-imatrix overview
Model name: MN-12B-Lyra-v4 Brief description: A finetune of Mistral Nemo by Sao10K. Uses the ChatML prompt format. Presets: You can use the built in ChatML presets within SillyTavern and adjust from there. Alternatively, check out Virt-io's ChatML v1.9 presets here, make sure you read the repository page for how to use them properly. Request page: https://huggingface.co/Lewdiculous/Model-Requests/discussions/75 Model link: https://huggingface.co/Sao10K/MN-12B-Lyra-v4 Quantized with llama.cpp: b3707 !image/png
Repository Files & Downloads
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| MN-12B-Lyra-v4-BF16.gguf | GGUF | BF16 | 22.82 GB | Download |
| MN-12B-Lyra-v4-F16.gguf | GGUF | F16 | 22.82 GB | Download |
| MN-12B-Lyra-v4-IQ2_S-imat.gguf | GGUF | IQ2_S | 3.85 GB | Download |
| MN-12B-Lyra-v4-IQ2_XXS-imat.gguf | GGUF | IQ2_XXS | 3.35 GB | Download |
| MN-12B-Lyra-v4-IQ3_M-imat.gguf | GGUF | IQ3_M | 5.33 GB | Download |
| MN-12B-Lyra-v4-IQ3_S-imat.gguf | GGUF | IQ3_S | 5.18 GB | Download |
| MN-12B-Lyra-v4-IQ3_XS-imat.gguf | GGUF | IQ3_XS | 4.94 GB | Download |
| MN-12B-Lyra-v4-IQ3_XXS-imat.gguf | GGUF | IQ3_XXS | 4.61 GB | Download |
| MN-12B-Lyra-v4-IQ4_XS-imat.gguf | GGUF | IQ4_XS | 6.28 GB | Download |
| MN-12B-Lyra-v4-Q3_K_L-imat.gguf | GGUF | Q3_K_L | 6.11 GB | Download |
| MN-12B-Lyra-v4-Q3_K_M-imat.gguf | GGUF | Q3_K_M | 5.67 GB | Download |
| MN-12B-Lyra-v4-Q4_K_M-imat.gguf | GGUF | Q4_K_M | 6.96 GB | Download |
| MN-12B-Lyra-v4-Q4_K_S-imat.gguf | GGUF | Q4_K_S | 6.63 GB | Download |
| MN-12B-Lyra-v4-Q5_K_M-imat.gguf | GGUF | Q5_K_M | 8.13 GB | Download |
| MN-12B-Lyra-v4-Q5_K_S-imat.gguf | GGUF | Q5_K_S | 7.93 GB | Download |
| MN-12B-Lyra-v4-Q6_K-imat.gguf | GGUF | Q6_K | 9.37 GB | Download |
| MN-12B-Lyra-v4-Q8_0-imat.gguf | GGUF | — | 12.13 GB | Download |
Model Details Live
Metadata Inspector
Normalized metadata (stored in metadata_json)
{
"metadata": {},
"card_data": {
"base_model": "Sao10K/MN-12B-Lyra-v4",
"license": "cc-by-nc-4.0",
"inference": false,
"tags": [
"mistral",
"nemo",
"roleplay",
"sillytavern"
],
"frontmatter": {
"base_model": "Sao10K/MN-12B-Lyra-v4",
"license": "cc-by-nc-4.0",
"inference": "false",
"tags": [
"mistral",
"nemo",
"roleplay",
"sillytavern"
]
},
"hero_image_url": "https://cdn-uploads.huggingface.co/production/uploads/65d4cf2693a0a3744a27536c/dVoru83WOpwVjMlgZ_xhA.png",
"summary": "**Model name:** MN-12B-Lyra-v4 **Brief description:** A finetune of Mistral Nemo by Sao10K. Uses the ChatML prompt format. > [!TIP] > **Presets:** > You can use the built in ChatML presets within SillyTavern and adjust from there. > Alternatively, check out Virt-io's ChatML v1.9 presets here, make sure you read the repository page for how to use them properly. > [!NOTE] > Request page: > https://huggingface.co/Lewdiculous/Model-Requests/discussions/75 > > Model link: > https://huggingface.co/Sao10K/MN-12B-Lyra-v4 > > Quantized with llama.cpp: > b3707 !image/png",
"quick_links": [],
"benchmark_table_html": "",
"readme_markdown": "---\nbase_model: Sao10K/MN-12B-Lyra-v4\nlicense: cc-by-nc-4.0\ninference: false\ntags:\n- mistral\n- nemo\n- roleplay\n- sillytavern\n---\n\n**Model name:** <br>\nMN-12B-Lyra-v4\n\n**Brief description:** <br>\nA finetune of Mistral Nemo by [Sao10K](https://huggingface.co/Sao10K/). <br>\nUses the ChatML prompt format. <br>\n\n> [!TIP]\n> **Presets:** <br>\n> You can use the built in ChatML presets within SillyTavern and adjust from there. <br>\n> Alternatively, check out [Virt-io's ChatML v1.9 presets here](https://huggingface.co/Virt-io/SillyTavern-Presets/tree/main/Prompts/ChatML/v1.9), make sure you read the [repository page for how to use them properly](https://huggingface.co/Virt-io/SillyTavern-Presets/).\n\n> [!NOTE]\n> Request page: <br>\n> https://huggingface.co/Lewdiculous/Model-Requests/discussions/75\n>\n> Model link: <br>\n> https://huggingface.co/Sao10K/MN-12B-Lyra-v4\n>\n> Quantized with llama.cpp: <br>\n> [b3707](https://github.com/ggerganov/llama.cpp/releases/tag/b3707)\n\n\n\n",
"related_quantizations": []
},
"tags": [
"nemo",
"gguf",
"mistral",
"roleplay",
"sillytavern",
"base_model:Sao10K/MN-12B-Lyra-v4",
"base_model:quantized:Sao10K/MN-12B-Lyra-v4",
"license:cc-by-nc-4.0",
"region:us",
"conversational"
],
"likes": 40,
"downloads": 7227,
"gated": false,
"private": false,
"last_modified": "2024-09-10T08:02:23.000Z",
"created_at": "2024-09-09T14:13:25.000Z",
"pipeline_tag": "",
"library_name": "nemo"
}
Source payload excerpt (from Hugging Face API)
{
"_id": "66df0285d05219e329c178bc",
"id": "Lewdiculous/MN-12B-Lyra-v4-GGUF-IQ-Imatrix",
"modelId": "Lewdiculous/MN-12B-Lyra-v4-GGUF-IQ-Imatrix",
"sha": "dad0c53ae8b48bd131e919d6740bb4db365a8cc7",
"createdAt": "2024-09-09T14:13:25.000Z",
"lastModified": "2024-09-10T08:02:23.000Z",
"author": "Lewdiculous",
"downloads": 7227,
"likes": 40,
"gated": false,
"private": false,
"pipeline_tag": "",
"library_name": "nemo",
"siblings_count": 22
}