Burn Candle Backend

Deprecated: This backend is deprecated and will be removed in a future release. Please migrate to one of the actively maintained backends:

• CubeCL backends (CUDA, ROCm, Vulkan, Metal, WebGPU) for GPU acceleration
• Flex (burn-flex) for portable pure-Rust CPU execution (std, no_std, WASM)
• LibTorch (burn-tch) for a mature CPU/GPU backend

Burn Candle Backend

Deprecated: This crate is deprecated as of 0.21.0-pre.2 and will be removed in a future release. Please migrate to one of the actively maintained backends:

• CubeCL backends (CUDA, ROCm, Vulkan, Metal, WebGPU) for GPU acceleration
• Flex (burn-flex) for portable pure-Rust CPU execution (std, no_std, WASM)
• LibTorch (burn-tch) for a mature CPU/GPU backend

This crate provides a backend for Burn based on the Candle framework.

Feature Flags

• cuda - Cuda GPU device (NVIDIA only)
• accelerate - Accelerate framework (macOS only)