richarderkhov/sethuiyer_-_llama3.2-3b-anvita-instruct-gguf overview

{
  "metadata": {},
  "card_data": {
    "frontmatter": {},
    "hero_image_url": "",
    "summary": "This is a merge of pre-trained language models created using mergekit.",
    "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\nLlama3.2-3B-Anvita-Instruct - GGUF\n- Model creator: https://huggingface.co/sethuiyer/\n- Original model: https://huggingface.co/sethuiyer/Llama3.2-3B-Anvita-Instruct/\n\n\n| Name | Quant method | Size |\n| ---- | ---- | ---- |\n| [Llama3.2-3B-Anvita-Instruct.Q2_K.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q2_K.gguf) | Q2_K | 1.39GB |\n| [Llama3.2-3B-Anvita-Instruct.IQ3_XS.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.IQ3_XS.gguf) | IQ3_XS | 1.53GB |\n| [Llama3.2-3B-Anvita-Instruct.IQ3_S.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.IQ3_S.gguf) | IQ3_S | 1.59GB |\n| [Llama3.2-3B-Anvita-Instruct.Q3_K_S.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q3_K_S.gguf) | Q3_K_S | 1.59GB |\n| [Llama3.2-3B-Anvita-Instruct.IQ3_M.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.IQ3_M.gguf) | IQ3_M | 1.65GB |\n| [Llama3.2-3B-Anvita-Instruct.Q3_K.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q3_K.gguf) | Q3_K | 1.73GB |\n| [Llama3.2-3B-Anvita-Instruct.Q3_K_M.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q3_K_M.gguf) | Q3_K_M | 1.73GB |\n| [Llama3.2-3B-Anvita-Instruct.Q3_K_L.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q3_K_L.gguf) | Q3_K_L | 1.85GB |\n| [Llama3.2-3B-Anvita-Instruct.IQ4_XS.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.IQ4_XS.gguf) | IQ4_XS | 1.91GB |\n| [Llama3.2-3B-Anvita-Instruct.Q4_0.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q4_0.gguf) | Q4_0 | 1.99GB |\n| [Llama3.2-3B-Anvita-Instruct.IQ4_NL.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.IQ4_NL.gguf) | IQ4_NL | 2.0GB |\n| [Llama3.2-3B-Anvita-Instruct.Q4_K_S.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q4_K_S.gguf) | Q4_K_S | 2.0GB |\n| [Llama3.2-3B-Anvita-Instruct.Q4_K.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q4_K.gguf) | Q4_K | 2.09GB |\n| [Llama3.2-3B-Anvita-Instruct.Q4_K_M.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q4_K_M.gguf) | Q4_K_M | 2.09GB |\n| [Llama3.2-3B-Anvita-Instruct.Q4_1.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q4_1.gguf) | Q4_1 | 2.18GB |\n| [Llama3.2-3B-Anvita-Instruct.Q5_0.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q5_0.gguf) | Q5_0 | 2.37GB |\n| [Llama3.2-3B-Anvita-Instruct.Q5_K_S.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q5_K_S.gguf) | Q5_K_S | 2.37GB |\n| [Llama3.2-3B-Anvita-Instruct.Q5_K.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q5_K.gguf) | Q5_K | 2.41GB |\n| [Llama3.2-3B-Anvita-Instruct.Q5_K_M.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q5_K_M.gguf) | Q5_K_M | 2.41GB |\n| [Llama3.2-3B-Anvita-Instruct.Q5_1.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q5_1.gguf) | Q5_1 | 2.55GB |\n| [Llama3.2-3B-Anvita-Instruct.Q6_K.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q6_K.gguf) | Q6_K | 2.76GB |\n| [Llama3.2-3B-Anvita-Instruct.Q8_0.gguf](https://huggingface.co/RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf/blob/main/Llama3.2-3B-Anvita-Instruct.Q8_0.gguf) | Q8_0 | 3.58GB |\n\n\n\n\nOriginal model description:\n---\nbase_model:\n- chuanli11/Llama-3.2-3B-Instruct-uncensored\n- Joycean0301/Llama-3.2-3B-Instruct-Medical-Conversational\nlibrary_name: transformers\ntags:\n- mergekit\n- merge\n\n---\n# merge\n\nThis is a merge of pre-trained language models created using [mergekit](https://github.com/cg123/mergekit).\n\n## Merge Details\n### Merge Method\n\nThis model was merged using the [DARE](https://arxiv.org/abs/2311.03099) [TIES](https://arxiv.org/abs/2306.01708) merge method using [chuanli11/Llama-3.2-3B-Instruct-uncensored](https://huggingface.co/chuanli11/Llama-3.2-3B-Instruct-uncensored) as a base.\n\n### Models Merged\n\nThe following models were included in the merge:\n* ./anvita\n* [Joycean0301/Llama-3.2-3B-Instruct-Medical-Conversational](https://huggingface.co/Joycean0301/Llama-3.2-3B-Instruct-Medical-Conversational)\n\n### Configuration\n\nThe following YAML configuration was used to produce this model:\n\n```yaml\n\nslices:\nmodels:\n  - model: Joycean0301/Llama-3.2-3B-Instruct-Medical-Conversational\n    parameters:\n      weight: [0.25, 0.35, 0.45, 0.35, 0.25]\n      density: [0.1, 0.25, 0.5, 0.25, 0.1]\n  - model: chuanli11/Llama-3.2-3B-Instruct-uncensored\n  - model: ./anvita\n    parameters:\n      weight: [0.55, 0.45, 0.35, 0.45, 0.55]\n      density: [0.1, 0.25, 0.5, 0.25, 0.1]\nmerge_method: dare_ties\nbase_model: chuanli11/Llama-3.2-3B-Instruct-uncensored\nparameters:\n  int8_mask: true\ndtype: bfloat16\n\n```\n\n## How to run\n```python\nfrom transformers import AutoTokenizer, AutoModelForCausalLM\nimport torch\n\n# Load the tokenizer\ntokenizer = AutoTokenizer.from_pretrained('./merge')\n\n# Load the model (safetensors)\nmodel = AutoModelForCausalLM.from_pretrained(\n    'sethuiyer/Llama3.2-3B-Anvita-Instruct',\n    torch_dtype=torch.bfloat16\n)\nmodel.to(\"cuda\")\nmodel.eval()\n\ndef generate_response(user_prompt):\n    input_ids = tokenizer(user_prompt, return_tensors=\"pt\").input_ids\n    input_ids = input_ids.to(\"cuda\")\n\n    # Generate response\n    with torch.no_grad():  # Disable gradient calculation for inference\n        output = model.generate(input_ids=input_ids, max_length=512)  # Adjust max_length as needed\n\n    # Decode the generated output\n    response = tokenizer.decode(output[0], skip_special_tokens=True)\n    return response\n\n# Example usage\nuser_prompt = \"<|start_header_id|>user<|end_header_id|>What is large language models?<|eot_id|><|start_header_id|>assistant<|end_header_id|>\"\nresponse = generate_response(user_prompt)\n\nresponse = response[response.index(\"assistant\") + len(\"assistant\"):]\nprint(\"Generated Response:\")\nprint(response)\n```\n\n\n\n\nAdditional thanks to @nicoboss for giving me access to his private supercomputer, enabling me to provide many more quants, at much higher speed, than I would otherwise be able to.",
    "related_quantizations": []
  },
  "tags": [
    "gguf",
    "arxiv:2311.03099",
    "arxiv:2306.01708",
    "endpoints_compatible",
    "region:us",
    "conversational"
  ],
  "likes": 0,
  "downloads": 212,
  "gated": false,
  "private": false,
  "last_modified": "2024-10-21T04:34:33.000Z",
  "created_at": "2024-10-21T04:01:59.000Z",
  "pipeline_tag": "",
  "library_name": ""
}

{
  "_id": "6715d237acae3e126d163f21",
  "id": "RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf",
  "modelId": "RichardErkhov/sethuiyer_-_Llama3.2-3B-Anvita-Instruct-gguf",
  "sha": "978970ae0b11c4dfa559e24d4f01bdf8c658a2dd",
  "createdAt": "2024-10-21T04:01:59.000Z",
  "lastModified": "2024-10-21T04:34:33.000Z",
  "author": "RichardErkhov",
  "downloads": 212,
  "likes": 0,
  "gated": false,
  "private": false,
  "pipeline_tag": "",
  "library_name": "",
  "siblings_count": 24
}