feat(model_cost): add display_name, model_vendor, and model_family to all models

[nmgarza5]

Title

Add display_name and model_vendor metadata to all models. Add model_version where relevant. These changes should not be breaking. They just add additional model info to the end user which provide a QoL boost. Parsing out the raw model name is painful as there is no standardized format.

some example updates:

    "512-x-512/dall-e-2": {
        "display_name": "DALL-E 2",
        "model_vendor": "openai",
        "input_cost_per_pixel": 6.86e-08,
        "litellm_provider": "openai",
        "mode": "image_generation",
        "output_cost_per_pixel": 0.0
    },
    "512-x-512/max-steps/stability.stable-diffusion-xl-v0": {
        "display_name": "Stable Diffusion XL",
        "litellm_provider": "bedrock",
        "max_input_tokens": 77,
        "max_tokens": 77,
        "mode": "image_generation",
        "model_vendor": "stability",
        "model_version": "v0",
        "output_cost_per_image": 0.036
    },
    "ai21.j2-mid-v1": {
        "display_name": "Jurassic-2 Mid",
        "input_cost_per_token": 1.25e-05,
        "litellm_provider": "bedrock",
        "max_input_tokens": 8191,
        "max_output_tokens": 8191,
        "max_tokens": 8191,
        "mode": "chat",
        "model_vendor": "ai21",
        "model_version": "v1",
        "output_cost_per_token": 1.25e-05
    },
    "ai21.j2-ultra-v1": {
        "display_name": "Jurassic-2 Ultra",
        "input_cost_per_token": 1.88e-05,
        "litellm_provider": "bedrock",
        "max_input_tokens": 8191,
        "max_output_tokens": 8191,
        "max_tokens": 8191,
        "mode": "chat",
        "model_vendor": "ai21",
        "model_version": "v1",
        "output_cost_per_token": 1.88e-05
    },

Pre-Submission checklist

Please complete all items before asking a LiteLLM maintainer to review your PR

• I have Added testing in the tests/litellm/ directory, Adding at least 1 test is a hard requirement - see details
• I have added a screenshot of my new test passing locally
• My PR passes all unit tests on make test-unit
• My PR's scope is as isolated as possible, it only solves 1 specific problem

Type

🆕 New Feature

Changes

This PR adds three new metadata fields to every model entry in model_prices_and_context_window.json:

• display_name: A human-readable name for the model (e.g., "Claude Sonnet 3.5", "GPT-4o", "Gemini 2.0 Flash")
• model_vendor: The organization that created the model (e.g., "anthropic", "openai", "google", "meta")

These fields enable better UI display and categorization of models in applications built on LiteLLM, making it easier to:

• Show user-friendly model names in dropdowns and selectors
• Filter and group models by vendor
• Build model selection interfaces with proper categorization

The metadata was generated using pattern matching against model keys and provider information to accurately detect vendors and families across 2000+ model entries.

[vercel]

@nmgarza5 is attempting to deploy a commit to the CLERKIEAI Team on Vercel.

A member of the Team first needs to authorize it.

[CLAassistant]

All committers have signed the CLA.

[vercel]

The latest updates on your projects. Learn more about Vercel for GitHub.

Project	Deployment	Preview	Comments	Updated (UTC)
litellm	Ready	Preview	Comment	Dec 3, 2025 1:23am

[nmgarza5]

Lint failures here are unrelated to these changes
https://github.com/BerriAI/litellm/actions/runs/19878203207/job/56970352510?pr=17330