no-llm 0.1.4

Standard Interface for Large Language Models

no/llm

Standard Interface for Large Language Models

---

no/llm is a Python library that provides a unified interface for working with LLMs, with built-in support for model configuration, parameter validation, and provider management.

[!IMPORTANT] ⚠️ Early Stage Development
This project is in early stages and under active development. While we're working hard to maintain stability, APIs and features may change as we improve the library. We encourage you to try it out and provide feedback, but please be aware that production use should be carefully considered.

Quick Install

uv pip install "no_llm[pydantic-ai]"

Quick Example with Pydantic AI

[!TIP] Get a free API key from OpenRouter to test various models without individual provider accounts.

import os

from no_llm.integrations.pydantic_ai import NoLLMModel
from no_llm.registry import ModelRegistry
from no_llm.settings import ValidationMode, settings
from pydantic_ai import Agent
from pydantic_ai.settings import ModelSettings

settings.validation_mode = ValidationMode.ERROR
# or ValidationMode.WARN, ValidationMode.CLAMP

os.environ["OPENROUTER_API_KEY"] = "..."

# Get model from registry
registry = ModelRegistry()
openrouter_models = list(registry.list_models(provider="openrouter"))
print([m.identity.id for m in openrouter_models])
# > ['claude-3.5-haiku', 'claude-3.5-sonnet-v2', 'claude-3.7-sonnet', 'deepseek-chat', 'deepseek-r1-llama-70b-distilled', 'deepseek-reasoner', ...]
no_llm_model = NoLLMModel(*openrouter_models)

# Use with Pydantic AI
agent = Agent(no_llm_model, model_settings=ModelSettings(temperature=1.2))
result = agent.run_sync("What is the capital of France?")
print(result.data)
# > 2025-04-09 09:50:51.375 | WARNING  | no_llm.integrations.pydantic_ai:request:220 - Model deepseek-chat failed, trying next fallback. Error: Invalid value for parameter 'temperature'
# > Current value: 1.2
# > Valid range: (0.0, 1.0)
# > Error: Value 1.2 outside range [0.0, 1.0]
# > 2025-04-09 09:50:51.375 | WARNING  | no_llm.integrations.pydantic_ai:request:220 - Model deepseek-r1-llama-70b-distilled failed, trying next fallback. Error: Invalid value for parameter 'temperature'
# > Current value: 1.2
# > Valid range: (0.0, 1.0)
# > Error: Value 1.2 outside range [0.0, 1.0]
# ✅ gemini-2.0-flash
# > The capital of France is **Paris**.

Why no/llm?

• Provider Agnostic: Support for OpenAI, Anthropic, Google, Mistral, Groq, and more through a single interface

• Built-in Validation: Type-safe parameter validation and capability checking

• Provider Fallbacks: Automatic fallback between providers and data centers

• Configuration System: YAML-based model configurations with inheritance support

• Model Registry: Central management of models with capability-based filtering

• Integration Ready: Works with Pydantic AI, and more frameworks coming soon

Next Steps

• Configuration Guide
• Parameter System
• Provider Documentation
• Registry System