Instructor
Python library designed for getting structured, type-safe data and Pydantic models from LLMs.
Explore, compare, and discover vetted AI tools, products, and experiments.
Python library designed for getting structured, type-safe data and Pydantic models from LLMs.
Unified proxy server and SDK to call 100+ LLMs using the standard OpenAI specification.
Ultra-fast open-source LLM fine-tuning engine providing 5x faster training with 80% less VRAM usage.
Declarative framework streamlining post-training and fine-tuning of leading open-weight LLMs.
PyTorch-native library for easily authoring, fine-tuning, and evaluating generative LLMs.
Production-grade open-source LLM evaluation framework like Pytest for LLM applications.
Stanford's framework for algorithmically optimizing LM prompts and weights instead of manual prompt engineering.
Fully managed data parsing, indexing, and retrieval platform for production LLM knowledge pipelines.
Open-source AI observability platform with OpenTelemetry-native tracing for LLM applications.
Universal memory layer for AI agents and assistants that learns and updates user preferences continuously.
The platform for building stateful AI agents with persistent, self-editing memory systems.
Modular open-source Python framework for building custom NLP and generative search pipelines.