← Back to Vibe Nano Framework

Framework Layer

Frameworks are where researchers and production teams spend most of their time. AGI progress depends on frameworks that expose power without hiding critical system trade-offs.

Framework Projects

1 of 4 projects complete

Nano Train

Done

Learning-first distributed LLM training framework built around Megatron-style parallelism, covering TP, PP, EP, DP, ZeRO-1/2, and mixed precision on a DeepSeek-style model stack.

github.com/lastweek/nano-train ↗
→

Nano Torch

In Progress

Deep learning framework with PyTorch-like API.

→

Nano Serving

TBD

Model serving and inference optimization system.

→

Nano Agentic RL

TBD

An agentic RL post-training framework.

→