Lucid3.0
DocsAPI ReferenceChangelog
Tensorlucid.Tensor88
Tensor Creation23
Tensor Operations158
Autogradlucid.autograd18
Distributionslucid.distributions57
Einopslucid.einops4
FFTlucid.fft22
Functional Transformslucid.func9
Linear Algebralucid.linalg39
Mixed Precisionlucid.amp1
Neural Networkslucid.nn168
Optimizerslucid.optim29
Profilerlucid.profiler6
OpEventCProfileSummaryCMemoryStatsCProfilerCprofilefrecord_functionf
Serializationlucid.serialization6
Signallucid.signal12
Special Functionslucid.special38
module

profiler

6 members
lucid.profiler

lucid.profiler: performance profiling utilities.

Classes

classOpEvent9 methods

A single recorded operation event captured by the profiler.

classProfileSummary6 methods

Aggregated timing summary for a single op name across many calls.

classMemoryStats6 methods

Snapshot of allocator state at a single point in time.

classProfiler10 methods

Record op-level timing and memory traces from the C++ engine.

Functions

fnprofile→ Iterator[Profiler]

Context manager that records ops executed within the `with` block.

fnrecord_function→ Iterator[None]

Annotate an enclosing region with a name visible to the profiler.

Lucid3.0

Production-grade ML framework for Apple Silicon. MLX + Accelerate native backend.

Documentation

  • Quickstart
  • Installation
  • Autograd
  • Metal Device

API Reference

  • lucid.Tensor
  • lucid.nn
  • lucid.optim
  • lucid.autograd

Resources

  • GitHub(opens in new tab)
  • Changelog

© 2026 Lucid. Built for Apple Silicon.

Python 3.14+ · macOS arm64