torchtune
PyTorch-native recipes you can actually read and modify.
90
sovereigntytorchtune is PyTorch's own fine-tuning library, built as readable, hackable training recipes rather than a framework with a configuration language on top. Nothing is hidden behind abstraction layers, which makes it the right choice when you need to change how training works rather than what it trains on — custom loss functions, unusual data pipelines, research variations. It is maintained inside the PyTorch project, so its dependency story is unusually clean.
OPEN SOURCEBSD-3-ClauseSELF-HOSTLOCAL-FIRST
LicenseBSD-3-Clause
PricingFree. Part of the PyTorch ecosystem.
Open sourceYes
Self-hostableYes
Local-first dataYes
What it does well
- +Recipes are plain PyTorch — readable and modifiable end to end
- +Maintained within the PyTorch project itself
- +Minimal dependency surface compared with the wrappers
- +BSD-3-Clause
Where it falls short
- −Fewer batteries included — you write more yourself
- −Smaller library of ready-made community configs
- −Assumes real PyTorch familiarity
torchtune as an alternative to
Where torchtune shows up in our comparisons, and how it ranked.
torchtune head-to-head
Straight comparisons against the tools people weigh it against.