https://github.com/ggerganov/llama.cpp/pull/7499 was added which adds --metadata metadata override feature so you can add extra metadata to the gguf file. The metadata override files may look like below (Note that // c style comment is not present in the actual file, but is there for illustrative purpose)
{
// Example Metadata Override Fields
"general.name" : "Example Model Six",
"general.author" : "John Smith",
"general.version" : "v1.0",
"general.organization" : "SparkExampleMind",
"general.quantized_by" : "Abbety Jenson",
"general.description" : "This is an example of a model",
// Useful for cleanly regenerating default naming conventions
"general.finetune" : "instruct",
"general.basename" : "llamabase",
"general.size_label" : "8x2.3Q",
// Licensing details
"general.license" : "apache-2.0",
"general.license.name" : "Apache License Version 2.0, January 2004",
"general.license.link" : "https://huggingface.co/datasets/choosealicense/licenses/blob/main/markdown/apache-2.0.md",
// Typically represents the converted GGUF repo (Unless native)
"general.url" : "https://huggingface.co/SparkExampleMind/llamabase-8x2.3Q-instruct-v1.0-F16/blob/main/README.md",
"general.doi" : "doi:10.1080/02626667.2018.1560449",
"general.uuid" : "f18383df-ceb9-4ef3-b929-77e4dc64787c",
"general.repo_url" : "https://huggingface.co/SparkExampleMind/llamabase-8x2.3Q-instruct-v1.0-F16",
// Model Source during conversion
"general.source.url" : "https://huggingface.co/SparkExampleMind/llamabase-8x2.3Q-instruct-v1.0-safetensor/blob/main/README.md",
"general.source.doi" : "doi:10.1080/02626667.2018.1560449",
"general.source.uuid" : "a72998bf-3b84-4ff4-91c6-7a6b780507bc",
"general.source.repo_url": "https://huggingface.co/SparkExampleMind/llamabase-8x2.3Q-instruct-v1.0-safetensor",
// Model Parents (Merges, Pre-tuning, etc...)
"general.base_models" : [
{
"name": "GPT-3",
"author": "OpenAI",
"version": "3.0",
"organization": "OpenAI",
"description": "A large language model capable of performing a wide variety of language tasks.",
"url": "https://openai.com/research/gpt-3",
"doi": "10.5555/gpt3doi123456",
"uuid": "123e4567-e89b-12d3-a456-426614174000",
"repo_url": "https://github.com/openai/gpt-3"
},
{
"name": "BERT",
"author": "Google AI Language",
"version": "1.0",
"organization": "Google",
"description": "A transformer-based model pretrained on English to achieve state-of-the-art performance on a range of NLP tasks.",
"url": "https://github.com/google-research/bert",
"doi": "10.5555/bertdoi789012",
"uuid": "987e6543-e21a-43f3-a356-527614173999",
"repo_url": "https://github.com/google-research/bert"
}
],
// Model Datasets Used (Training data...)
"general.datasets": [
{
"name": "Wikipedia Corpus",
"author": "Wikimedia Foundation",
"version": "2021-06",
"organization": "Wikimedia",
"description": "A dataset comprising the full English Wikipedia, used to train models in a range of natural language tasks.",
"url": "https://dumps.wikimedia.org/enwiki/",
"doi": "10.5555/wikidoi234567",
"uuid": "234e5678-f90a-12d3-c567-426614172345",
"repo_url": "https://github.com/wikimedia/wikipedia-corpus"
},
{
"name": "Common Crawl",
"author": "Common Crawl Foundation",
"version": "2021-04",
"organization": "Common Crawl",
"description": "A dataset containing web-crawled data from various domains, providing a broad range of text.",
"url": "https://commoncrawl.org",
"doi": "10.5555/ccdoi345678",
"uuid": "345e6789-f90b-34d5-d678-426614173456",
"repo_url": "https://github.com/commoncrawl/cc-crawl-data"
}
],
// Array Based Metadata
"general.tags": ["text generation", "transformer", "llama", "tiny", "tiny model"],
"general.languages": ["en"]
}
