CGo wrapper for mlx-c providing zero-Python Metal GPU inference. Includes Gemma 3 model architecture, BPE tokenizer, KV cache, composable sampling, and OpenAI-compatible serve command. Build-tagged (darwin && arm64 && mlx) with stubs for cross-platform. Co-Authored-By: Claude Opus 4.6 <noreply@anthropic.com>
10 lines
425 B
Go
10 lines
425 B
Go
//go:build !(darwin && arm64 && mlx)
|
|
|
|
// Package mlx provides Go bindings for Apple's MLX framework via mlx-c.
|
|
// This stub file is used on non-darwin/non-arm64 platforms or when the
|
|
// mlx build tag is not set. All operations report MLX as unavailable.
|
|
package mlx
|
|
|
|
// MetalAvailable reports whether Metal GPU is available.
|
|
// Always returns false on non-Apple Silicon platforms.
|
|
func MetalAvailable() bool { return false }
|