Labels

• arch: amd64

  arch: amd64

  x86-64 specific (AVX2, SIMD, future)

• arch: arm64

  arch: arm64

  ARM64 specific (Apple Silicon, NEON)

• arch: wasm

  arch: wasm

  WebAssembly support

• area: api

  area: api

  Public API surface (tensor/, nn/, optim/)

• area: autodiff

  area: autodiff

  Automatic differentiation, gradient tape

• area: cli

  area: cli

  CLI tools (born, born-bench, born-convert)

• area: cpu

  area: cpu

  CPU backend, element-wise ops, BLAS

• area: examples

  area: examples

  Example projects (MNIST, CNN, GPU)

• area: generate

  area: generate

  Text generation, LLM inference pipeline

• area: gguf

  area: gguf

  GGUF loader, K-quant dequantization

• area: nn

  area: nn

  NN modules (Linear, Conv2D, RMSNorm, Embedding)

• area: onnx

  area: onnx

  ONNX model import, 30+ operators

• area: optim

  area: optim

  Optimizers (SGD, Adam)

• area: serialization

  area: serialization

  Model save/load (.born format), mmap

• area: tensor

  area: tensor

  Core tensor ops, shapes, dtypes, Backend interface

• area: tokenizer

  area: tokenizer

  Tokenizers (TikToken, BPE)

• area: webgpu

  area: webgpu

  WebGPU/GPU backend, WGSL shaders

• bench: vs-burn

  bench: vs-burn

  Comparison with Burn (Rust)

• bench: vs-onnxrt

  bench: vs-onnxrt

  Comparison with ONNX Runtime

• bench: vs-pytorch

  bench: vs-pytorch

  Comparison with PyTorch

• effort: 1

  effort: 1

  Trivial, < 1 hour

• effort: 13

  effort: 13

  Epic, 2+ weeks (split)

• effort: 2

  effort: 2

  Small, 1-4 hours

• effort: 3

  effort: 3

  Medium, ~1 day

• effort: 5

  effort: 5

  Large, 2-3 days

• effort: 8

  effort: 8

  Very large, ~1 week

• good first issue

  good first issue

  Good for newcomers

• help wanted

  help wanted

  Extra attention needed

• mentor available

  mentor available

  Maintainer will mentor

• model: born

  model: born

  Native .born format