outlines

OpenRLHF is a high-performance RLHF training framework for fine-tuning large language models (7B-70B+ parameters) using methods like PPO, DPO, and GRPO. It leverages Ray for distributed architecture and vLLM for accelerated inference, achieving speeds 2x faster than alternatives like DeepSpeedChat. Use this skill when you need efficient, distributed RLHF training with optimized GPU resource sharing and ZeRO-3 support.

SGLang is a high-performance LLM serving framework that uses RadixAttention for automatic prefix caching, enabling significantly faster structured generation. It's ideal for developers needing JSON/regex outputs, constrained decoding, or building agentic workflows with tool calls. Use it when you require up to 5× faster inference than alternatives like vLLM in scenarios with shared prefixes.

DSPy is a framework for building complex AI systems like RAG pipelines and agents using declarative programming. It automatically optimizes prompts and LM calls based on your data, moving beyond manual prompt engineering. Use it to create modular, maintainable, and systematically improved AI applications.

Instructor is a structured output library that extracts validated data from LLM responses using Pydantic schemas. It automatically retries failed extractions and provides type-safe JSON parsing with streaming support. Use it when you need reliable, validated data extraction from LLMs like OpenAI or Anthropic.