Agent Reinforcement Trainer: train multi-step agents for real-world tasks using GRPO. Give your agents on-the-job training. Reinforcement learning for Qwen3.6, GPT-OSS, Llama, and more!

Tool for generating high quality Synthetic datasets

Learning to Retrieve by Trying - Source code for Grounding by Trying: LLMs with Reinforcement Learning-Enhanced Retrieval

AgentFlow: In-the-Flow Agentic System Optimization

RL-driven framework that composes modular DSPy pipelines and teleprompters to improve LLM reasoning (experiments use GPT-2).

GEMBA — GPT Estimation Metric Based Assessment

dspy-cli is a tool for creating, developing, testing, and deploying DSPy programs as HTTP APIs.

A production framework for DSPy implementing the Teacher-Student pattern. Distill the reasoning of expensive models (Teacher) into optimized prompts for cheap, fast models (Student) to reduce infer…

Demonstration of DSPy optimization for Skill.md files

RVAA: Recursive Vision-Action Agent for Long Video Understanding. Implementation of the RLM paradigm (Zhang, Kraska, Khattab 2025)

A DSPy integration for the Agent Skills specification, enabling ReAct agents to dynamically discover and use modular, sandboxed skills

This Project for Creating unified tool to generate synthetic data (text and sensorial data) using LLM with the help of DSPy

Hypernetworks that adapt LLMs for specific benchmark tasks using only textual task description as the input

The pretty much "official" DSPy framework for Typescript