LLM Providers

Configure OpenAI, Anthropic, and Gemini

Provider Comparison

Provider	Models	Best For
OpenAI	gpt-4o, gpt-4o-mini, gpt-4-turbo	General purpose, cost-effective
Anthropic	claude-sonnet-4, claude-3-opus	Advanced reasoning, long context
Gemini	gemini-1.5-pro, gemini-pro	Multimodal, Google ecosystem

---

Configuration

OpenAI

Environment:

export OPENAI_API_KEY=sk-...

Config:

from echo import LLMConfig, get_llm

llm_config = LLMConfig(
    provider="openai",
    model="gpt-4o-mini",
    temperature=0.2,
    max_tokens=2000,
)

llm = get_llm(llm_config)

Models: gpt-4o, gpt-4o-mini, gpt-4-turbo, gpt-3.5-turbo

Anthropic

Environment:

export ANTHROPIC_API_KEY=sk-ant-...

Config:

from echo import LLMConfig, get_llm

llm_config = LLMConfig(
    provider="anthropic",
    model="claude-sonnet-4-20250514",
    temperature=0.2,
    max_tokens=2000,
)

llm = get_llm(llm_config)

Models: claude-sonnet-4-20250514, claude-3-opus-20240229, claude-3-haiku-20240307

Gemini

Environment:

export GOOGLE_API_KEY=...

Config:

from echo import LLMConfig, get_llm

llm_config = LLMConfig(
    provider="gemini",
    model="gemini-1.5-pro",
    temperature=0.2,
    max_tokens=2000,
)

llm = get_llm(llm_config)

Models: gemini-1.5-pro, gemini-pro, gemini-pro-vision

---

Switching Providers

Change one line to switch providers:

# OpenAI
llm_config = LLMConfig(provider="openai", model="gpt-4o-mini")

# Anthropic
llm_config = LLMConfig(provider="anthropic", model="claude-sonnet-4-20250514")

# Gemini
llm_config = LLMConfig(provider="gemini", model="gemini-1.5-pro")

Agent code remains the same.

---

LLMConfig Parameters

Parameter	Type	Default	Description
provider	str	Required	"openai", "anthropic", "gemini"
model	str	Required	Model identifier
temperature	float	0.7	Randomness (0.0-1.0)
max_tokens	int	4096	Max response length
max_iterations	int	5	Max tool use iterations

Next Steps

Streaming

Real-time responses

Examples

Complete examples