Apple AI models you can actually run

APPLE MODELS · DEMOS · GUIDES

Apple’s research teams publish far more than FastVLM. This hub organises the models by task, marks which ones have an in-browser tool or a tested local guide on this site, and links every entry to its official source.

Vision

Vision-language models, image–text encoders and depth estimation that run on-device.

[### FastVLM

Browser tool

Vision-language model with the FastViTHD encoder; 0.5B runs in your browser.

Released 2025Open](/blog/fastvlm-models)[### Depth Pro

Browser tool

Metric depth and focal length from one image; powers the depth map generator.

Released 2024Open](/blog/depth-pro)[### MobileCLIP2

Browser tool

Low-latency image–text encoders; zero-shot classification and similarity tools.

Released 2025Open](/blog/mobileclip2)[### AIMv2

Run guide

Open vision encoders (0.3B–2.7B) with PyTorch, JAX and MLX backends.

Released 2024Open](/blog/aimv2)

3D

Single-image 3D: nearby-view synthesis and relightable asset generation.

[### SHARP

Run guide

One photo to 3D Gaussians and real-time nearby views in under a second.

Released 2025Open](/blog/apple-sharp-image-to-3d)[### Luce

Reference

Relightable PBR Gaussians and textured meshes from a single object image.

Released Aug 2026Open](/blog/luce)

Image & video generation

Normalizing-flow generators for images, video and unified multimodal generation.

[### STARFlow · STARFlow-V · STARFlow2

Reference

Scalable normalizing flows for images, video and unified multimodal generation.

Released 2025–2026Open](/blog/starflow)

Language & code

Research models for code generation and retrieval-augmented reasoning.

[### DiffuCoder

Reference

7B masked-diffusion language model for code generation.

Released 2025Open](/blog/apple-diffucoder)[### CLaRa

Reference

Retrieval and generation joined through continuous latent reasoning.

Released 2025Open](/blog/apple-clara)[### SimpleSD

Reference

Embarrassingly simple self-distillation that improves code generation (Apple Research).

Released 2026Official page](https://machinelearning.apple.com/research/simple-self-distillation)

Apple Intelligence

The Apple Foundation Models behind Apple Intelligence and the developer framework that exposes them.

[### Apple Foundation Models (AFM 3)

Run guide

AFM 3 Core, Core Advanced and the Private Cloud Compute models behind Apple Intelligence.

Released Jun 2026Open](/blog/apple-foundation-models)[### Foundation Models framework

Run guide

Swift and Python APIs for on-device LLM sessions, guided generation, tools and multimodal prompts.

Released WWDC25 · WWDC26Open](/blog/apple-foundation-models-framework)

What the badges mean

• Browser tool: runs in your browser on this site, no install.
• Run guide: local setup steps verified against the official repository.
• Reference: research summary with official links; no runnable release yet or out of scope for on-device use.

Start with a task instead

Depth maps, classification, similarity and image Q&A—each tool names the Apple model behind it.

Browse tools

Beyond Apple

Compare FastVLM with SmolVLM2, LFM2.5-VL, Gemma and other small VLMs that run on-device.

On-device VLM hub