Pydantic Ai Model Integration

This walks you through wiring up different LLM providers in PydanticAI using simple string formats like `openai:gpt-4o` or `anthropic:claude-sonnet-4-5`. The fallback model setup is the standout feature: chain multiple providers so if one hits rate limits or goes down, it automatically tries the next. You also get streaming for both plain text and structured Pydantic output, per-run model settings overrides, and usage tracking that shows token counts across requests. Useful when you're building something that needs to stay up despite API hiccups or when you want to route cheaper requests to mini models and save the expensive ones for complex work.

skills.sh installs: 120
GitHub stars: 87
last updated: Aug 21, 2026

View on GitHub

Install

$ npx -y skills add existential-birds/beagle --skill pydantic-ai-model-integration --agent claude-code

Installs into .claude/skills of the current project.

Files

PydanticAI Model Integration

Provider Model Strings

Format: provider:model-name

from pydantic_ai import Agent

# OpenAI
Agent('openai:gpt-4o')
Agent('openai:gpt-4o-mini')
Agent('openai:o1-preview')

# Anthropic
Agent('anthropic:claude-sonnet-4-5')
Agent('anthropic:claude-haiku-4-5')

# Google (API Key)
Agent('google-gla:gemini-2.0-flash')
Agent('google-gla:gemini-2.0-pro')

# Google (Vertex AI)
Agent('google-vertex:gemini-2.0-flash')

# Groq
Agent('groq:llama-3.3-70b-versatile')
Agent('groq:mixtral-8x7b-32768')

# Mistral
Agent('mistral:mistral-large-latest')

# Other providers
Agent('cohere:command-r-plus')
Agent('bedrock:anthropic.claude-3-sonnet')

Model Settings

from pydantic_ai import Agent
from pydantic_ai.settings import ModelSettings

agent = Agent(
    'openai:gpt-4o',
    model_settings=ModelSettings(
        temperature=0.7,
        max_tokens=1000,
        top_p=0.9,
        timeout=30.0,  # Request timeout
    )
)

# Override per-run
result = await agent.run(
    'Generate creative text',
    model_settings=ModelSettings(temperature=1.0)
)