AIPython★ 204
whitecircle/halo
HALO: FRAMEWORK FOR LLM TRAINING
Halo is an open-source framework from White Circle for training large language and multimodal models. It supports agent-based training, async-RL, distributed training, and fine-tuning via GRPO. Written in Python, it has garnered 204 stars on GitHub.
// KEY FEATURES
- Trains LLM and multimodal models using the GRPO algorithm.
- Supports agent-based training with environment and policy integration.
- Enables async-RL parallel agent training in a distributed cluster.
- Allows fine-tuning large models on multiple GPUs with automatic sharding.
#agent#async-rl#distributed-training#fine-tuning#grpo#large-language-models#llm#lora#model-parallelism#moe#multimodal#peft
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