AIPython★ 202
Birfy/agentdescent
AGENT DESCENT: EVOLUTION OF AGENTS
AgentDescent — framework where model parameters are agents that learn via gradient descent. Their differentials are gradients, aggregator is optimizer; it supports parallel asynchronous launch and self-training of skills, prompts, harnesses.
// KEY FEATURES
- Launches hundreds of agents in parallel asynchronous mode with isolated state.
- Computes gradients as diffs between current and new agent versions.
- Aggregator combines gradients and applies the optimizer to skills, prompts, harnesses.
- Skills, prompts, and harnesses of agents improve via gradient descent.
#agent-framework#agents#ai-agents#asynchronous#evolutionary-algorithms#gradient-descent#llm#llm-agents#optimizer#parallel#prompt-optimization#self-evolving-agents
Open on GitHub →New repositories every 30 minutes
REDDYX AI scans GitHub 24/7 and ships the best AI/ML/Web3 projects to Telegram.
Join on Telegram