Flow Nexus Neural

This is a full-featured neural network training platform that runs inside distributed E2B sandboxes. You get single-node training for standard architectures (feedforward, LSTM, transformer, GAN) with configurable layer stacks, plus distributed training clusters that can coordinate multiple sandboxes with different topologies (mesh, ring, star). The template marketplace lets you deploy pre-trained models instead of building from scratch. Federated learning support keeps data on local nodes, which matters for privacy-sensitive work like medical imaging. The distributed setup handles parameter servers, workers, and gradient aggregation automatically. If you need to train models that don't fit on one machine or want to experiment with distributed architectures without managing infrastructure, this handles the orchestration.