Alibaba-Quark/SSP
Search Self-Play: Pushing the Frontier of Agent Capability without Supervision
This project helps machine learning engineers and researchers train advanced AI agents that can solve complex questions by intelligently using search engines. It takes a starting set of prompts for problem generation and produces a powerful AI agent capable of multi-turn reasoning and information retrieval. The project is designed for those who want to develop highly capable AI for knowledge-intensive tasks without needing vast amounts of human-labeled data.
Use this if you are building an AI agent that needs to answer complex, real-world questions by dynamically searching for information, and you want to train it efficiently without extensive human-annotated examples.
Not ideal if your AI agent's task does not require external search capabilities or multi-step reasoning, or if you prefer traditional supervised learning methods with readily available labeled datasets.
Stars
97
Forks
8
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 04, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/Alibaba-Quark/SSP"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
ai4co/reevo
[NeurIPS 2024] ReEvo: Large Language Models as Hyper-Heuristics with Reflective Evolution
SALT-NLP/collaborative-gym
Framework and toolkits for building and evaluating collaborative agents that can work together...
Gen-Verse/LatentMAS
Latent Collaboration in Multi-Agent Systems
lean-dojo/LeanCopilot
LLMs as Copilots for Theorem Proving in Lean
WooooDyy/AgentGym-RL
Code and implementations for the paper "AgentGym-RL: Training LLM Agents for Long-Horizon...