Model Intelligence Sheet
richarderkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf overview
---
Downloads
131
Likes
0
Pipeline
—
Library
—
Visibility
Public
Access
Open
Repository Files & Downloads
22 files detected
Direct downloads for all repository files
| File | Type | Quantization | Size | Link |
|---|---|---|---|---|
| marathi-gpt-gemma-2b.IQ3_M.gguf | GGUF | IQ3_M | 1.22 GB | Download |
| marathi-gpt-gemma-2b.IQ3_S.gguf | GGUF | IQ3_S | 1.20 GB | Download |
| marathi-gpt-gemma-2b.IQ3_XS.gguf | GGUF | IQ3_XS | 1.16 GB | Download |
| marathi-gpt-gemma-2b.IQ4_NL.gguf | GGUF | IQ4_NL | 1.45 GB | Download |
| marathi-gpt-gemma-2b.IQ4_XS.gguf | GGUF | IQ4_XS | 1.40 GB | Download |
| marathi-gpt-gemma-2b.Q2_K.gguf | GGUF | Q2_K | 1.08 GB | Download |
| marathi-gpt-gemma-2b.Q3_K.gguf | GGUF | Q3_K | 1.29 GB | Download |
| marathi-gpt-gemma-2b.Q3_K_L.gguf | GGUF | Q3_K_L | 1.36 GB | Download |
| marathi-gpt-gemma-2b.Q3_K_M.gguf | GGUF | Q3_K_M | 1.29 GB | Download |
| marathi-gpt-gemma-2b.Q3_K_S.gguf | GGUF | Q3_K_S | 1.20 GB | Download |
| marathi-gpt-gemma-2b.Q4_0.gguf | GGUF | — | 1.44 GB | Download |
| marathi-gpt-gemma-2b.Q4_1.gguf | GGUF | — | 1.56 GB | Download |
| marathi-gpt-gemma-2b.Q4_K.gguf | GGUF | Q4_K | 1.52 GB | Download |
| marathi-gpt-gemma-2b.Q4_K_M.gguf | GGUF | Q4_K_M | 1.52 GB | Download |
| marathi-gpt-gemma-2b.Q4_K_S.gguf | GGUF | Q4_K_S | 1.45 GB | Download |
| marathi-gpt-gemma-2b.Q5_0.gguf | GGUF | — | 1.68 GB | Download |
| marathi-gpt-gemma-2b.Q5_1.gguf | GGUF | — | 1.79 GB | Download |
| marathi-gpt-gemma-2b.Q5_K.gguf | GGUF | Q5_K | 1.71 GB | Download |
| marathi-gpt-gemma-2b.Q5_K_M.gguf | GGUF | Q5_K_M | 1.71 GB | Download |
| marathi-gpt-gemma-2b.Q5_K_S.gguf | GGUF | Q5_K_S | 1.68 GB | Download |
| marathi-gpt-gemma-2b.Q6_K.gguf | GGUF | Q6_K | 1.92 GB | Download |
| marathi-gpt-gemma-2b.Q8_0.gguf | GGUF | — | 2.49 GB | Download |
Model Details Live
Metadata Inspector
Normalized metadata (stored in metadata_json)
{
"metadata": {},
"card_data": {
"frontmatter": {},
"hero_image_url": "",
"summary": "---",
"quick_links": [],
"benchmark_table_html": "",
"readme_markdown": "Quantization made by Richard Erkhov.\n\n[Github](https://github.com/RichardErkhov)\n\n[Discord](https://discord.gg/pvy7H8DZMG)\n\n[Request more models](https://github.com/RichardErkhov/quant_request)\n\n\nmarathi-gpt-gemma-2b - GGUF\n- Model creator: https://huggingface.co/l3cube-pune/\n- Original model: https://huggingface.co/l3cube-pune/marathi-gpt-gemma-2b/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [marathi-gpt-gemma-2b.Q2_K.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q2_K.gguf) | Q2_K | 1.08GB |\n| [marathi-gpt-gemma-2b.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.IQ3_XS.gguf) | IQ3_XS | 1.16GB |\n| [marathi-gpt-gemma-2b.IQ3_S.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.IQ3_S.gguf) | IQ3_S | 1.2GB |\n| [marathi-gpt-gemma-2b.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q3_K_S.gguf) | Q3_K_S | 1.2GB |\n| [marathi-gpt-gemma-2b.IQ3_M.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.IQ3_M.gguf) | IQ3_M | 1.22GB |\n| [marathi-gpt-gemma-2b.Q3_K.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q3_K.gguf) | Q3_K | 1.29GB |\n| [marathi-gpt-gemma-2b.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q3_K_M.gguf) | Q3_K_M | 1.29GB |\n| [marathi-gpt-gemma-2b.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q3_K_L.gguf) | Q3_K_L | 1.36GB |\n| [marathi-gpt-gemma-2b.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.IQ4_XS.gguf) | IQ4_XS | 1.4GB |\n| [marathi-gpt-gemma-2b.Q4_0.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q4_0.gguf) | Q4_0 | 1.44GB |\n| [marathi-gpt-gemma-2b.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.IQ4_NL.gguf) | IQ4_NL | 1.45GB |\n| [marathi-gpt-gemma-2b.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q4_K_S.gguf) | Q4_K_S | 1.45GB |\n| [marathi-gpt-gemma-2b.Q4_K.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q4_K.gguf) | Q4_K | 1.52GB |\n| [marathi-gpt-gemma-2b.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q4_K_M.gguf) | Q4_K_M | 1.52GB |\n| [marathi-gpt-gemma-2b.Q4_1.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q4_1.gguf) | Q4_1 | 1.56GB |\n| [marathi-gpt-gemma-2b.Q5_0.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q5_0.gguf) | Q5_0 | 1.68GB |\n| [marathi-gpt-gemma-2b.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q5_K_S.gguf) | Q5_K_S | 1.68GB |\n| [marathi-gpt-gemma-2b.Q5_K.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q5_K.gguf) | Q5_K | 1.71GB |\n| [marathi-gpt-gemma-2b.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q5_K_M.gguf) | Q5_K_M | 1.71GB |\n| [marathi-gpt-gemma-2b.Q5_1.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q5_1.gguf) | Q5_1 | 1.79GB |\n| [marathi-gpt-gemma-2b.Q6_K.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q6_K.gguf) | Q6_K | 1.92GB |\n| [marathi-gpt-gemma-2b.Q8_0.gguf](https://huggingface.co/RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf/blob/main/marathi-gpt-gemma-2b.Q8_0.gguf) | Q8_0 | 2.49GB |\n\n\n\n\nOriginal model description:\n---\nlicense: cc-by-4.0\nlanguage: mr\nwidget:\n# - text: <bos>\\n### Instruction:\\n(9+0)+(10+5)? 3 चरणांमध्ये सोडवा\\n\\n### Input:\\n\\n\\n### Response:\\n\n- text: <bos>\\n### Instruction:\\nमहाराष्ट्राची राजधानी काय आहे?\\n\\n### Input:\\n\\n\\n### Response:\\n\n---\n\n## MahaGemma-2B\nMahaGemma-2B is a Marathi Gemma model. It is a Gemma 2B (google/gemma-2b) model LoRA fine-tuned on translated Marathi datasets. \n[dataset link] (https://github.com/l3cube-pune/MarathiNLP)\n\nThis is part of the MahaNLP initiative. More details coming soon. <br>\n\nPrompt format:\n```\n<bos>\\n### Instruction:\\nमहाराष्ट्राची राजधानी काय आहे?\\n\\n### Input:\\n\\n\\n### Response:\\nमहाराष्ट्राची राजधानी मुंबई आहे\n```\n\nCiting\n```\n@article{joshi2022l3cube,\n title={L3cube-mahanlp: Marathi natural language processing datasets, models, and library},\n author={Joshi, Raviraj},\n journal={arXiv preprint arXiv:2205.14728},\n year={2022}\n}\n\n```\nModel Family: <br>\n<a href=\"https://huggingface.co/l3cube-pune/marathi-gpt-gemma-2b\"> MahaGemma-2B </a> <br>\n<a href=\"https://huggingface.co/l3cube-pune/marathi-gpt-gemma-7b\"> MahaGemma-7B </a>\n\n",
"related_quantizations": []
},
"tags": [
"gguf",
"arxiv:2205.14728",
"endpoints_compatible",
"region:us"
],
"likes": 0,
"downloads": 131,
"gated": false,
"private": false,
"last_modified": "2024-06-22T19:06:03.000Z",
"created_at": "2024-06-22T18:43:16.000Z",
"pipeline_tag": "",
"library_name": ""
}
Source payload excerpt (from Hugging Face API)
{
"_id": "66771b44d0970f1efac86b25",
"id": "RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf",
"modelId": "RichardErkhov/l3cube-pune_-_marathi-gpt-gemma-2b-gguf",
"sha": "2b8c6e1496627511b1a27738ce2230a84005d5f1",
"createdAt": "2024-06-22T18:43:16.000Z",
"lastModified": "2024-06-22T19:06:03.000Z",
"author": "RichardErkhov",
"downloads": 131,
"likes": 0,
"gated": false,
"private": false,
"pipeline_tag": "",
"library_name": "",
"siblings_count": 24
}