I am working on an app using FoundationModels to process web pages. I am looking to find ways to filter the input to fit within the token limits. I have unit tests, UI tests and the app running on an iPad in the simulator. It appears that the different configurations of the test environment seems to affect the token limits. That is, the same input in a unit test and UI test will hit different token limits. Is this correct? Or is this an artifact of my test tooling?

Hi, I'm looking for the best way to use MLX models, particularly those I've fine-tuned, within a React Native application on iOS devices. Is there a recommended integration path or specific API for bridging MLX's capabilities to React Native for deployment on iPhones and iPads?

Is anything configurable for LanguageModelSession.Guardrails besides the default? I'm prototyping a camping app, and it's constantly slamming into guardrail errors when I use the new foundation model interface. Any subjects relating to fishing, survival, etc. won't generate. For example the prompt "How can I kill deer ticks using a clothing treatment?" returns a generation error. The results that I get are great when it works, but so far the local model sessions are extremely unreliable.

Hello, I was trying to test out Foundation Model however it says Model assets are unavailable. I got my MacBook M1 back in China when i was living there. is this due to region lock?

Introduced in the Keynote was the 3D Lock Screen images with the kangaroo: https://9to5mac.com/wp-content/uploads/sites/6/2025/06/3d-lock-screen-2.gif I can't see any mention on if this effect is available for developers with an API to convert flat 2D photos in to the same 3D feeling image. Does anyone know if there is an API?

Hi all. My adapter model just won't invoke my tool. The problem I am having is covered in an older post: https://developer.apple.com/forums/thread/794839?answerId=852262022#852262022 Sadly the thread dies there and no resolution is seen in that thread. It's worth noting that I have developed an AI chatbot built around LanguageModelSession to which I feed the exact same system prompt that I feed to my training set (pasted further in this post). The AI chatbot works perfectly, the tool is invoked when needed. I am training the adapter model because the base model whilst capable doesn't produce the quality I'm looking for. So here's the template of an item in my training set: [ { 'role': 'system', 'content': systemPrompt, 'tools': [TOOL_DEFINITION] }, { 'role': 'user', 'content': entry['prompt'] }, { 'role': 'assistant', 'content': entry['code'] } ] where TOOL_DEFINITION = { 'type': 'function', 'function': { 'name': 'WriteUbersichtWidgetToFileSystem', 'description': 'Writes an Übersicht Widget to the file system. Call this tool as the last step in processing a prompt that generates a widget.', 'parameters': { 'type': 'object', 'properties': { 'jsxContent': { 'type': 'string', 'description': 'Complete JSX code for an Übersicht widget. This should include all required exports: command, refreshFrequency, render, and className. The JSX should be a complete, valid Übersicht widget file.' } }, 'required': ['jsxContent'] } } ... and systemPrompt = A conversation between a user and a helpful assistant. You are an Übersicht widget designer. Create Übersicht widgets when requested by the user. IMPORTANT: You have access to a tool called WriteUbersichtWidgetToFileSystem. When asked to create a widget, you MUST call this tool. ### Tool Usage: Call WriteUbersichtWidgetToFileSystem with complete JSX code that implements the Übersicht Widget API. Generate custom JSX based on the user's specific request - do not copy the example below. ### Übersicht Widget API (REQUIRED): Every Übersicht widget MUST export these 4 items: - export const command: The bash command to execute (string) - export const refreshFrequency: Refresh rate in milliseconds (number) - export const render: React component function that receives {output} prop (function) - export const className: CSS positioning for absolute placement (string) Example format (customize for each request): WriteUbersichtWidgetToFileSystem({jsxContent: `export const command = "echo hello"; export const refreshFrequency = 1000; export const render = ({output}) => { return <div>{output}</div>; }; export const className = "top: 20px; left: 20px;"`}) ### Rules: - The terms "ubersicht widget", "widget", "a widget", "the widget" must all be interpreted as "Übersicht widget" - Generate complete, valid JSX code that follows the Übersicht widget API - When you generate a widget, don't just show JSON or code - you MUST call the WriteUbersichtWidgetToFileSystem tool - Report the results to the user after calling the tool ### Examples: - "Generate a Übersicht widget" → Use WriteUbersichtWidgetToFileSystem tool - "Can you add a widget that shows the time" → Use WriteUbersichtWidgetToFileSystem tool - "Create a widget with a button" → Use WriteUbersichtWidgetToFileSystem tool When the script that I use to compose the full training set is executed, entry['prompt'] and entry['code'] contain the prompt and the resulting JSX code for one of the examples I'm feeding to the training session. This is repeated for about 60 such examples that I have in my sample data collection. Thanks for any help. Michael

I didn't run benchmarks before update, but it seems at least 5x slower. Of course all the LLM work is on remote servers, so is non-intuitive to me this should be happening. Had updated MacOS and Xcode to 26.1RC at the same time, so can't even say I think it is MacOS or I think it is Xcode. Before the update the progress indicator for each piece of code might seem to get stuck at the very end (and toggling between Navigators and Coding Assistant) in Xcode UI seemed to refresh the UI and confirm coding complete... but now it seems progress races to 50%, then often is stuck at 75%... well earlier than used to get stuck. And it like something is legitimately processing not just a UI glitch. I'm wondering if this is somehow tied to visual rendering of the code in the little white window? CMD-TAB into Xcode seems laggy. Xcode is pinning a CPU. Why, this is all remote LLM work? MacBook Pro 2021 M1 64GB RAM. Went from 26.01 to 26.1RC. Didn't touch any of the betas until RC1.

Hi, I have been trying to integrate a CoreML model into Xcode. The model was made using tensorflow layers. I have included both the model info and a link to the app repository. I am mainly just really confused on why its not working. It seems to only be printing the result for case 1 (there are 4 cases labled, case 0, case 1, case 2, and case 3). If someone could help work me through this error that would be great! here is the link to the repository: https://github.com/ShivenKhurana1/Detect-to-Protect-App this file with the model code is called SecondView.swift and here is the model info: Input: conv2d_input-> image (color 224x224) Output: Identity -> MultiArray (Float32 1x4)

When I ran the following code on a physical iPhone device that supports Apple Intelligence, I encountered the following error log. What does this internal error code mean? Image generation failed with NSError in a different domain: Error Domain=ImagePlaygroundInternal.ImageGeneration.GenerationError Code=11 “(null)”, returning a generic error instead let imageCreator = try await ImageCreator() let style = imageCreator.availableStyles.first ?? .animation let stream = imageCreator.images(for: [.text("cat")], style: style, limit: 1) for try await result in stream { // error: ImagePlayground.ImageCreator.Error.creationFailed _ = result.cgImage }

After watching the What's new in App Intents session I'm attempting to create an intent conforming to URLRepresentableIntent. The video states that so long as my AppEntity conforms to URLRepresentableEntity I should not have to provide a perform method . My application will be launched automatically and passed the appropriate URL. This seems to work in that my application is launched and is passed a URL, but the URL is in the form: FeatureEntity/{id}. Am I missing something, or is there a trick that enables it to pass along the URL specified in the AppEntity itself? struct MyExampleIntent: OpenIntent, URLRepresentableIntent { static let title: LocalizedStringResource = "Open Feature" static var parameterSummary: some ParameterSummary { Summary("Open \(\.$target)") } @Parameter(title: "My feature", description: "The feature to open.") var target: FeatureEntity } struct FeatureEntity: AppEntity { // ... } extension FeatureEntity: URLRepresentableEntity { static var urlRepresentation: URLRepresentation { "https://myurl.com/\(.id)" } }

I am trying to benchmark and see if the Qwen3 1.7B model can run in an iPhone SE 3 [4 GB RAM]. My core problem is - Even with weight quantization the SE 3 is not able to load into memory. What I've tried: I am converting a Torch model to the Core ML format using coremltools. I have tried the following combinations of quantization and context length 8 bit + 1024 8 bit + 2048 4 bit + 1024 4 bit + 2048 All the above quantizations are done with dynamic shape with the default being [1,1] in the hope that the whole context length does not get allocated in memory The 4-bit model is approximately 865MB on disk The 8-bit model is approximately 1.7 GB on disk During load: With the int4 quantization the memory spikes during intitial load a lot. Could this be because many operations are converted to int8 or fp16 as core ML does not perform operations natively on int4? With int8 on the profiler the memory does not go above 2 GB (only 900 MB) but it is still not able to load as it shows the following error. 2GB is the limit where jetsam kills the app for the iPhone SE 3 E5RT: Error(s) occurred compiling MIL to BNNS graph: [CreateBnnsGraphProgramFromMIL]: BNNS Graph Compile: failed to preallocate file with error: No space left on device for path: /var/mobile/Containers/Data/Application/ 5B8BB7D2-06A6-4BAE-A042-407B6D805E7C/Library/Caches /com.tss.qwen3-coreml/ com.apple.e5rt.e5bundlecache/ 23A341/<long key>.tmp.12586_4362093968.bundle/ H14.bundle/main/main_bnns/bnns_program.bnnsir Some online sources have suggested activation quantization but I am unsure if that will have any impact on loading [as the spike is during load and not inference] The model spec also suggests that there is no dequantization happening (for e.g from 4 bit -> fp16) So I had couple of queries: Has anyone faced similar issues? What could be the reasons for the temporary memory spike during LOAD What are approaches that can be adopted to deal with this issue? Any help would be greatly appreciated. Thank you.

HI, I've been modifying the Camera sample app found here: https://developer.apple.com/tutorials/sample-apps/capturingphotos-camerapreview ... in the processpreview images, I am calling in to the Vision APis to either detect a person or object, then I'm using the segmentation mask to extract the person and composite them onto a different background with some other filters. I am using coreimage to filter the CIImages, and converting and displaying as a SwiftUI Image. When running on my IPhone, it works fine. When running on my Iphone with the debugger, it crashes within a few seconds... Attached is a screenshot. At the top is an EXC_BAD_ACCESS in libRPAC.dylib`std::__1::__hash_table<std::__1::__hash_value_type<long, qos_info_t>, std::__1::__unordered_map_hasher<long, std::__1::__hash_value_type<long, qos_info_t>, std::__1::hash, std::__1::equal_to, true>, std::__1::__unordered_map_equal<long, std::__1::__hash_value_type<long, qos_info_t>, std::__1::equal_to, std::__1::hash, true>, std::__1::allocator<std::__1::__hash_value_type<long, qos_info_t>>>::__emplace_unique_key_args<long, std::__1::piecewise_construct_t const&, std::__1::tuple<long const&>, std::__1::tuple<>>: This was working fine a couple of days ago.. Not sure why it's popping up now. Am I correct in interpreting this as an LLDB issue? How do I fix it?

Hi, unfortunately I am not able to verify this but I remember some time ago I was able to create CoreML models that had one (or more) inputs with an enumerated shape size, and one (or more) inputs with a static shape. This was some months ago. Since then I updated my MacOS to Sequoia 15.5, and when I try to execute MLModels with this setup I get the following error libc++abi: terminating due to uncaught exception of type CoreML::MLNeuralNetworkUtilities::AsymmetricalEnumeratedShapesException: A model doesn't allow input features with enumerated flexibility to have unequal number of enumerated shapes, but input feature global_write_indices has 1 enumerated shapes and input feature input_hidden_states has 3 enumerated shapes. It may make sense (but not really though) to verify that for inputs with a flexible enumerated shape they all have the same number of possible shapes is the same, but this should not impede the possibility of also having static shape inputs with a single shape defined alongside the flexible shape inputs.

About the Core ML model encryption mention in:https://developer.apple.com/documentation/coreml/encrypting-a-model-in-your-app Ｗhen I encrypted the model, if the machine is M chip, the model will load perfectly. One the other hand, when I test the executable on an Intel chip macbook, there will be an error: Error Domain=com.apple.CoreML Code=9 "Operation not supported on this platform." UserInfo={NSLocalizedDescription=Operation not supported on this platform.} Intel test machine is 2019 macbook air with CPU: Intel i5-8210Y, OS: 14.7.6 23H626, With Apple T2 Security Chip. The encrypted model do load on M2 and M4 macbook air. If the model is NOT encrypted, it will also load on the Intel test machine. I did not find in Core ML document that suggest if the encryption/decryption support Intel chips. May I check if the decryption indeed does NOT support Intel chip?

We are trying to implement a custom encryption scheme for our Core ML models. Our goal is to bundle encrypted models, decrypt them into memory at runtime, and instantiate the MLModel without the unencrypted model file ever touching the disk. We have looked into the native apple encryption described here https://developer.apple.com/documentation/coreml/encrypting-a-model-in-your-app but it has limitations like not working on intel macs, without SIP, and doesn’t work loading from dylib. It seems like most of the Core ML APIs require a file path, there is MLModelAsset APIs but I think they just write a modelc back to disk when compiling but can’t find any information confirming that (also concerned that this seems to be an older API, and means we need to compile at runtime). I am aware that the native encryption will be much more secure but would like not to have the models in readable text on disk. Does anyone know if this is possible or any alternatives to try to obfuscate the Core ML models, thanks

When using Foundation Models, is it possible to ask the model to produce output in a specific language, apart from giving an instruction like "Provide answers in ." ? (I tried that and it kind of worked, but it seems fragile.) I haven't noticed an API to do so and have a use-case where the output should be in a user-selectable language that is not the current system language.

Greetings! I was trying to get a response from the LanguageModelSession but I just keep getting the following: Error getting response: Model Catalog error: Error Domain=com.apple.UnifiedAssetFramework Code=5000 "There are no underlying assets (neither atomic instance nor asset roots) for consistency token for asset set com.apple.MobileAsset.UAF.FM.Overrides" UserInfo={NSLocalizedFailureReason=There are no underlying assets (neither atomic instance nor asset roots) for consistency token for asset set com.apple.MobileAsset.UAF.FM.Overrides} This occurs both in macOS 15.5 running the new Xcode beta with an iOS 26 simulator, and also on a macOS 26 with Xcode beta. The simulators are both Pro iPhone 16s. I was wondering if anyone had any advice?

I am writing an app that parses text and conducts some actions. I don't want to give too much away ;) However, I am having a huge problem with token sizes. LanguageModelSession will of course give me the on device model 4096 available, but when you go over 4096, my code doesn't seem to be falling back to PCC, or even the system configured ChatGPT. Can anyone assist me with this? For some reason, after reading the docs, it's very unclear how this transition between the three takes place.

When trying to open an encrypted CoreML model file on a system with SIP disabled, the error message is Failed to generate key request for <...> with error: -42187 This should state that SIP is disabled and needs to be enabled.

At WWDC25 we launched a new type of Lab event for the developer community - Group Labs. A Group Lab is a panel Q&A designed for a large audience of developers. Group Labs are a unique opportunity for the community to submit questions directly to a panel of Apple engineers and designers. Here are the highlights from the WWDC25 Group Lab for Machine Learning and AI Frameworks. What are you most excited about in the Foundation Models framework? The Foundation Models framework provides access to an on-device Large Language Model (LLM), enabling entirely on-device processing for intelligent features. This allows you to build features such as personalized search suggestions and dynamic NPC generation in games. The combination of guided generation and streaming capabilities is particularly exciting for creating delightful animations and features with reliable output. The seamless integration with SwiftUI and the new design material Liquid Glass is also a major advantage. When should I still bring my own LLM via CoreML? It's generally recommended to first explore Apple's built-in system models and APIs, including the Foundation Models framework, as they are highly optimized for Apple devices and cover a wide range of use cases. However, Core ML is still valuable if you need more control or choice over the specific model being deployed, such as customizing existing system models or augmenting prompts. Core ML provides the tools to get these models on-device, but you are responsible for model distribution and updates. Should I migrate PyTorch code to MLX? MLX is an open-source, general-purpose machine learning framework designed for Apple Silicon from the ground up. It offers a familiar API, similar to PyTorch, and supports C, C++, Python, and Swift. MLX emphasizes unified memory, a key feature of Apple Silicon hardware, which can improve performance. It's recommended to try MLX and see if its programming model and features better suit your application's needs. MLX shines when working with state-of-the-art, larger models. Can I test Foundation Models in Xcode simulator or device? Yes, you can use the Xcode simulator to test Foundation Models use cases. However, your Mac must be running macOS Tahoe. You can test on a physical iPhone running iOS 18 by connecting it to your Mac and running Playgrounds or live previews directly on the device. Which on-device models will be supported? any open source models? The Foundation Models framework currently supports Apple's first-party models only. This allows for platform-wide optimizations, improving battery life and reducing latency. While Core ML can be used to integrate open-source models, it's generally recommended to first explore the built-in system models and APIs provided by Apple, including those in the Vision, Natural Language, and Speech frameworks, as they are highly optimized for Apple devices. For frontier models, MLX can run very large models. How often will the Foundational Model be updated? How do we test for stability when the model is updated? The Foundation Model will be updated in sync with operating system updates. You can test your app against new model versions during the beta period by downloading the beta OS and running your app. It is highly recommended to create an "eval set" of golden prompts and responses to evaluate the performance of your features as the model changes or as you tweak your prompts. Report any unsatisfactory or satisfactory cases using Feedback Assistant. Which on-device model/API can I use to extract text data from images such as: nutrition labels, ingredient lists, cashier receipts, etc? Thank you. The Vision framework offers the RecognizeDocumentRequest which is specifically designed for these use cases. It not only recognizes text in images but also provides the structure of the document, such as rows in a receipt or the layout of a nutrition label. It can also identify data like phone numbers, addresses, and prices. What is the context window for the model? What are max tokens in and max tokens out? The context window for the Foundation Model is 4,096 tokens. The split between input and output tokens is flexible. For example, if you input 4,000 tokens, you'll have 96 tokens remaining for the output. The API takes in text, converting it to tokens under the hood. When estimating token count, a good rule of thumb is 3-4 characters per token for languages like English, and 1 character per token for languages like Japanese or Chinese. Handle potential errors gracefully by asking for shorter prompts or starting a new session if the token limit is exceeded. Is there a rate limit for Foundation Models API that is limited by power or temperature condition on the iPhone? Yes, there are rate limits, particularly when your app is in the background. A budget is allocated for background app usage, but exceeding it will result in rate-limiting errors. In the foreground, there is no rate limit unless the device is under heavy load (e.g., camera open, game mode). The system dynamically balances performance, battery life, and thermal conditions, which can affect the token throughput. Use appropriate quality of service settings for your tasks (e.g., background priority for background work) to help the system manage resources effectively. Do the foundation models support languages other than English? Yes, the on-device Foundation Model is multilingual and supports all languages supported by Apple Intelligence. To get the model to output in a specific language, prompt it with instructions indicating the user's preferred language using the locale API (e.g., "The user's preferred language is en-US"). Putting the instructions in English, but then putting the user prompt in the desired output language is a recommended practice. Are larger server-based models available through Foundation Models? No, the Foundation Models API currently only provides access to the on-device Large Language Model at the core of Apple Intelligence. It does not support server-side models. On-device models are preferred for privacy and for performance reasons. Is it possible to run Retrieval-Augmented Generation (RAG) using the Foundation Models framework? Yes, it is possible to run RAG on-device, but the Foundation Models framework does not include a built-in embedding model. You'll need to use a separate database to store vectors and implement nearest neighbor or cosine distance searches. The Natural Language framework offers simple word and sentence embeddings that can be used. Consider using a combination of Foundation Models and Core ML, using Core ML for your embedding model.