model-discovery

AI model discovery fetches current model names directly from provider APIs like OpenAI, Anthropic, and Gemini. This retrieves up-to-date model identifiers to prevent errors from outdated references during API integration.

Classify AI models into performance tiers by categorizing fetched model names into 'fast', 'default', and 'heavy' groups. This tier classification enables dynamic API call routing based on task complexity and performance requirements.

A local Ollama instance is required only if you want to fetch and classify models from that specific provider. Network access to external provider APIs like Anthropic, OpenAI, and Gemini is required to retrieve their current model lists.

Detect newly released AI models by fetching current model names from provider APIs and comparing them against cached references. This new model detection alerts you to timely integrations for dynamic API workflows.

API calls fail with outdated AI model identifiers because providers frequently deprecate old model versions. Fetching real-time model names ensures you are always using the most appropriate and up-to-date models for your tasks.

You can use model discovery to find cost-effective models by fetching the latest 'fast' tier model IDs. This categorization helps select quick processing options or 'heavy' tier models for complex reasoning tasks.