Agent Model Resolution

How Agent() picks and routes an LLM when you omit llm= — from your environment

Agent() picks a model from your environment when you don’t pass llm=, then routes provider-prefixed defaults through litellm so a single provider key is enough to start.

As of PR #4795, the SDK Agent() constructor uses the same resolver as praisonai run — set one provider key and the Agent works with no llm=.

Quick Start

1. Simplest — one provider key

   Set a single provider credential and construct an Agent with no llm=:

   import os
   from praisonaiagents import Agent
   
   os.environ["OLLAMA_HOST"] = "http://localhost:11434"
   
   # No llm=, no OPENAI_API_KEY — the Agent picks ollama/llama3.2
   # and routes through litellm.
   agent = Agent(instructions="Answer briefly.")
   agent.start("2+2")

2. Explicit provider-prefixed

   Pass a provider/model string to route through litellm:

   from praisonaiagents import Agent
   
   agent = Agent(llm="anthropic/claude-3-5-sonnet-latest")
   agent.start("Write a haiku about the sea.")

3. Bare OpenAI model

   A bare model name uses the native OpenAI client:

   from praisonaiagents import Agent
   
   agent = Agent(llm="gpt-4o")
   agent.start("Summarise the plot of Hamlet.")

How It Works

The Agent scans your environment for a credential when llm= is omitted, then forks on whether the resolved model carries a provider/ prefix.

The first provider whose credential is present wins, in the order below. OpenAI is first so existing OpenAI users keep their default. The actual id is the capability-ranked best-available model for that provider at run time; the models below are shown as representatives — check praisonai models list --provider <p> for the current pick.

Resolution Table

The class an Agent uses depends on the shape of the resolved model — a provider/ prefix routes through litellm, a bare name uses the native OpenAI client.

Before PR #4795, six of the seven env-var defaults resolved the right model name but were then sent to the native OpenAI client — the Agent raised ValueError: OPENAI_API_KEY environment variable is required. Only the bare OpenAI default worked. The fix routes any provider-prefixed default through litellm, matching praisonai run.

Common Patterns

I only have a Groq key

import os
from praisonaiagents import Agent

os.environ["GROQ_API_KEY"] = "gsk-..."

# Picks groq/llama-3.3-70b-versatile, routed through litellm.
agent = Agent(instructions="Answer fast.")
agent.start("Name three prime numbers.")

I have an Ollama server running

import os
from praisonaiagents import Agent

os.environ["OLLAMA_HOST"] = "http://localhost:11434"

# Picks ollama/llama3.2, routed through litellm — no cloud key needed.
agent = Agent(instructions="Answer locally.")
agent.start("Explain black holes to a 5-year-old.")

Best Practices

Related