KwaiKEG/KwaiAgents
A generalized information-seeking agent system with Large Language Models (LLMs).
Combines Meta-agent tuning to train LLMs with planning, reflection, and tool-use capabilities across multiple base models (Qwen, Baichuan). Provides KAgentSys-Lite, a lightweight agent framework evolved from BabyAGI/Auto-GPT, alongside 200k+ instruction-tuning examples and 3,000+ evaluation benchmarks covering planning, tool-use, reflection, and profiling dimensions. Deployable via vLLM and FastChat with OpenAI-compatible API endpoints.
1,195 stars. No commits in the last 6 months.
Stars
1,195
Forks
114
Language
Python
License
—
Category
Last pushed
Jun 19, 2024
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/KwaiKEG/KwaiAgents"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
bubbuild/republic
Build LLM workflows like normal Python while keeping a full audit trail by default.
mitdbg/palimpzest
A System for Optimized Semantic Computation
ctrl-gaurav/effGen
effGen: Enabling Small Language Models as Capable Autonomous Agents
dlMARiA/Syzygy-of-thoughts
Syzygy-of-thoughts
lwcsrf/netflux
Minimalist framework for authoring custom agentic applications in python; emphasizes task...