Core AI Models 1.0.0

First stable release of Core AI Models: export recipes, Python primitives, and Swift runtime utilities for building on-device AI with Core AI.

Requirements: macOS and iOS 27.0+, Xcode 27.0+.

Highlights

• Model export: Recipes to export open-source models from Hugging Face and other sources to standalone .aimodel files. Run uv run coreai.model.registry --list-models to see what's supported.
• Compression presets for macOS and iOS:
• macOS: INT4 weight-only (default), and INT4 with INT8 KV cache.
• iOS: 4-bit palettization with channel group size 32 or 8.
• Custom recipes: You can pass your own coreai-opt YAML config.
• Python primitives: Building blocks for authoring custom Core AI models in PyTorch, with helpers for KV cache updates and forward methods.
• Swift runtime package (coreai-models): Libraries for language models (CoreAILM), image and video diffusion, image and video segmentation, speech, and object detection.
• Command line tools:
• Language models: llm-runner, llm-server, llm-benchmark
• Vision: image-segmenter, video-segmenter, object-detector
• Generation: diffusion-runner, videodiffusion-runner
• Speech: speech-recognizer
• Agent skills: A plugin for Claude Code, Codex CLI, and Gemini CLI with three skills: working-with-coreai, model-authoring, and model-compression-exploration.

Supported models