{
  "source_url": "https://huggingface.co/TinyLlama/TinyLlama-1.1B-Chat-v0.6/resolve/main/ggml-model-q4_0.gguf",
  "id": "tinyllama-1.1b",
  "object": "model",
  "name": "TinyLlama Chat 1.1B Q4",
  "version": "1.0",
  "description": "The TinyLlama project, featuring a 1.1B parameter Llama model, is pretrained on an expansive 3 trillion token dataset. Its design ensures easy integration with various Llama-based open-source projects. Despite its smaller size, it efficiently utilizes lower computational and memory resources, drawing on GPT-4's analytical prowess to enhance its conversational abilities and versatility.",
  "format": "gguf",
  "settings": {
      "ctx_len": 2048,
      "system_prompt": "<|system|>\n",
      "user_prompt": "<|user|>\n",
      "ai_prompt": "<|assistant|>\n"
  },
  "parameters": {
      "max_tokens": 2048
  },
  "metadata": {
      "author": "TinyLlama",
      "tags": ["Tiny", "Foundation Model"],
      "size": 637000000
  },
  "engine": "nitro"
}