Hugging Face ships AnyLanguageModel 1.0 for Apple platforms
Hugging Face has released AnyLanguageModel 1.0, alongside swift-huggingface 1.0 and swift-xet 1.0, a year after introducing the project. AnyLanguageModel is described as a drop-in replacement for Apple's Foundation Models framework that lets apps choose the model and provider, and the post says tool calling and guided generation now work with every provider, with image support using Foundation Models 27 syntax. Hosted support has expanded to OpenAI's Chat Completions and Responses APIs, Open Responses, Anthropic's Messages API with adaptive thinking and effort options, Google's Gemini API, Ollama, Private Cloud Compute on OS 27, and Hugging Face Inference Providers or other OpenAI-compatible endpoints via a custom base URL; the project also runs on Linux.
The project grew out of developer feedback from app developers curious about local models but unsure how to start and wanting to avoid supporting multiple inference providers, and the post argues local and hosted inference each have strengths with a shared API making it easier to choose the right model per task.
#AnyLanguageModel#Hugging Face#swift-huggingface#swift-xet#Apple Foundation Models#llama.cpp#MLX#Core ML#OpenAI#Anthropic#Google Gemini API#Ollama