Quick start

Basic Usage

from SimplerLLM.language.llm import LLM, LLMProvider

# Create an LLM instance
llm = LLM.create(provider=LLMProvider.OPENAI, model_name="gpt-4o")

# Generate text
response = llm.generate_response(prompt="What is Python?")
print(response)

Switching Providers

The power of SimplerLLM is that switching providers is a one-line change. The generate_response() method works the same way across all providers.

from SimplerLLM.language.llm import LLM, LLMProvider

# OpenAI
llm = LLM.create(provider=LLMProvider.OPENAI, model_name="gpt-4o")

# Anthropic Claude
llm = LLM.create(provider=LLMProvider.ANTHROPIC, model_name="claude-sonnet-4-5-20250929")

# Google Gemini
llm = LLM.create(provider=LLMProvider.GEMINI, model_name="gemini-2.0-flash")

# DeepSeek
llm = LLM.create(provider=LLMProvider.DEEPSEEK, model_name="deepseek-chat")

# Ollama (local)
llm = LLM.create(provider=LLMProvider.OLLAMA, model_name="llama3")

All providers use the same method:

response = llm.generate_response(prompt="Your prompt here")

Available Providers

Provider	Enum Value	Example Model
OpenAI	LLMProvider.OPENAI	gpt-4o
Anthropic	LLMProvider.ANTHROPIC	claude-sonnet-4-5-20250929
Google Gemini	LLMProvider.GEMINI	gemini-2.0-flash
DeepSeek	LLMProvider.DEEPSEEK	deepseek-chat
Cohere	LLMProvider.COHERE	command-r-plus
OpenRouter	LLMProvider.OPENROUTER	openai/gpt-4o
Perplexity	LLMProvider.PERPLEXITY	sonar
Ollama	LLMProvider.OLLAMA	llama3

Common Parameters

response = llm.generate_response(
    prompt="Explain quantum computing",
    system_prompt="You are a helpful assistant",
    temperature=0.7,
    max_tokens=500
)

Parameter	Description
prompt	The input text
system_prompt	Instructions for the model's behavior
temperature	Randomness (0.0 = deterministic, 1.0+ = creative)
max_tokens	Maximum length of the response

Vision / Image Input

Pass images for the model to analyze:

response = llm.generate_response(
    prompt="Describe this image",
    images=["https://example.com/photo.jpg"],
    detail="auto"  # "low", "high", or "auto"
)

Works with local files too:

response = llm.generate_response(
    prompt="What's in these images?",
    images=["path/to/image1.png", "path/to/image2.jpg"]
)

See LLM Interface for full details.

Getting Token Usage

Set full_response=True to get token counts and metadata:

response = llm.generate_response(
    prompt="Hello!",
    full_response=True
)

print(response.generated_text)
print(f"Input tokens: {response.input_token_count}")
print(f"Output tokens: {response.output_token_count}")