burn
v0.21.0 GrowingFlexible and Comprehensive Deep Learning Framework in Rust
Quick Verdict
- ✓Actively maintained (updated 67d ago)
- !Pre-1.0: API may have breaking changes
- ✓Massive adoption (2.0K crates depend on it)
- !Heavy dependency tree (23 direct deps)
- ✓Permissive license (MIT OR Apache-2.0)
Security
Deep Insights
118.2K downloads in the last 30 days (3.9K/day). Volume is roughly flat compared to the previous period.
2.0K crates depend on burn. Strong ecosystem adoption means battle-tested code and long-term stability.
The primary maintainer publishes 123 crates. This suggests deep Rust expertise and long-term commitment to the ecosystem.
Despite being 3+ years old, burn hasn't reached 1.0 yet. Expect potential API changes between versions.
23 direct dependencies. Consider the impact on compile times and supply chain complexity.
Notable dependents include fsrs, fibertools-rs, futuresdr, general-mcmc, bytecon. When high-quality crates choose burn, it's a strong quality signal.
Health Breakdown
Recency, release consistency, active ratio
Yanked ratio, deps, size, maturity, features
Reverse deps, ownership, ecosystem
Downloads, momentum, growth trend
Docs, repo, license, metadata
Download Trend
Top Dependents
Most downloaded crates that depend on burn
Version Adoption
Release Timeline
Feature Flags
default =["std", "burn-core/default", "burn-train?/default", "burn-collective?/default", "burn-candle?/default", "burn-cpu?/default", "burn-flex?/default", "burn-ndarray?/default", "burn-tch?/default", "burn-wgpu?/default", "burn-router?/default", "burn-remote?/default", "burn-cuda?/default", "burn-autodiff?/default", "burn-rocm?/default", "burn-nn/default", "burn-optim/default", "burn-dispatch?/default"]