llm7-validator 2025.6.110528

Validator for LLM7 chat completion requests with Pydantic and model ID constraints.

llm7-validator

llm7-validator is a Python package designed to validate message structures sent to LLM7-compatible chat completion APIs. It ensures that model names, messages, and attachments conform to the expected structure and size constraints.

Installation

To install llm7-validator, use pip:

pip install llm7-validator

Usage

Here is a minimal example of how to use it:

from llm7_validator import validate_chat_completion_request

check = validate_chat_completion_request({
    "model": "open-mistral-7b",
    "json_mode": True,
    "messages": [
        {"role": "system", "content": "hj"},
        {
            "role": "user",
            "content": [
                {"type": "text", "text": "What is this image?"},
                {"type": "image_url", "image_url": "https://example.com/image.png"},
            ],
        },
    ],
})

print(check)

Features

• Verifies message format (role, content, etc.)

• Validates model names against a dynamically fetched list from llm7-models.json

• Checks:

  • Image URLs have allowed extensions (.png, .jpg, etc.)
  • Base64 image content length
  • Total request size (max 5 MB)
  • Required fields and optional tuning parameters

• Uses cached HTTP responses for model list with a 5-minute expiration

Requirements

• Python ≥ 3.10
• pydantic==2.11.5
• requests-cache==1.2.1

Contributing

Contributions, issues, and feature requests are welcome! Please visit the GitHub repo to get started.

License

llm7-validator is licensed under the MIT License.