I’m keep looking around documentation and some sample codes but still haven’t found example of how was used this type of Network Regressor . Does it take some special parameters to perform on ANE , what size,format of DataFrame ?

I watched this year WWDC25 "Read Documents using the Vision framework". At the end of video there is mention of new DetectHandPoseRequest model for hand pose detection in Vision API. I looked Apple documentation and I don't see new revision. Moreover probably typo in video because there is only DetectHumanPoseRequst (swift based) and VNDetectHumanHandPoseRequest (obj-c based) (notice lack of Human prefix in WWDC video) First one have revision only added in iOS 18+: https://developer.apple.com/documentation/vision/detecthumanhandposerequest/revision-swift.enum/revision1 Second one have revision only added in iOS14+: https://developer.apple.com/documentation/vision/vndetecthumanhandposerequestrevision1 I don't see any new revision targeting iOS26+

Hi team, We have implemented a writing tool inside a WebView that allows users to type content in a textarea. When the "Show Writing Tools" button is clicked, an AI-powered editor opens. After clicking the "Rewrite" button, the AI modifies the text. However, when clicking the "Replace" button, the rewritten text does not update the original textarea. Kindly check and help me showButton.addTarget(self, action: #selector(showWritingTools(_:)), for: .touchUpInside) @available(iOS 18.2, *) optional func showWritingTools(_ sender: Any) Note: same cases working in TextView pfa

Does anyone know if ExecuTorch is officially supported or has been successfully used on visionOS? If so, are there any specific build instructions, example projects, or potential issues (like sandboxing or memory limitations) to be aware of when integrating it into an Xcode project for the Vision Pro? While ExecuTorch has support for iOS, I can't find any official documentation or community examples specifically mentioning visionOS. Thanks.

@property (assign,nonatomic) long long experimentalMLE5EngineUsage; //@synthesize experimentalMLE5EngineUsage=_experimentalMLE5EngineUsage - In the implementation block What is it, and why would disabling it fix NMS for a MLProgram? Is there anyway to signal this flag from model metadata? Is there anyway to signal or disable from a global, system-level scope? It's extremely easy to reproduce, but do not know how to investigate the drastic regression between toggling this flag let config = MLModelConfiguration() config.setValue(1, forKey: "experimentalMLE5EngineUsage")

The Core ML developer guide recommends saving reusable compiled Core ML models to a permanent location to avoid unnecessary rebuilds when creating a Core ML model instance. However, there is no location that remains consistent across app updates, since each update changes the UUID associated with the app’s resources path /var/mobile/Containers/Data/Application/<UUID>/Library/Application Support/ As a result, Core ML rebuilds models even if they are unchanged and located in the same relative directory within the app’s file structure.

Hi i'm curently crating a model to identify car plates (object detection) i use asitop to monitor my macbook pro and i see that only the cpu is used for the training and i wanted to know why

I'm experimenting with the new SpeechTranscriber in macOS/iOS 26, transcribing speech from a prerecorded mp4 file. Speed and quality are amazing! I've told the transcriber to include time indexes. Each run is always exactly one word, which can be very useful. When I look at the indexes the end of one run is always identical to the start of the next run, even if there's a pause. I'd like to identify pauses, perhaps to generate something like phrases for subtitling. With each run of text going into the next I can't do this, other than using punctuation - which might be rather rough. Any suggestions on detecting pauses, or getting that kind of metadata from the transcriber? Here's a short sample, showing each run with the start, end, and characters in the run: 105.9 --> 107.04 I 107.04 --> 107.16 think 107.16 --> 108.0 more 108.0 --> 108.42 lighting 108.42 --> 108.6 is 108.6 --> 108.72 definitely 108.72 --> 109.2 needed, 109.2 --> 109.92 downtown. 109.98 --> 110.4 My 110.4 --> 110.52 only 110.52 --> 110.7 question 110.7 --> 111.06 is, 111.06 --> 111.48 poll 111.48 --> 111.78 five, 111.78 --> 111.84 that 111.84 --> 112.08 you're 112.08 --> 112.38 increasing 112.38 --> 112.5 the 112.5 --> 113.34 50,000? 113.4 --> 113.58 Where 113.58 --> 113.88 exactly

I'm trying to set up Facebook AI's "Segment Anything" MLModel to compare its performance and efficacy on-device against the Vision library's Foreground Instance Mask Request. The Vision request accepts any reasonably-sized image for processing, and then has a method to produce an output at the same resolution as the input image. Conversely, the MLModel for Segment Anything accepts a 1024x1024 image for inference and outputs a 1024x1024 image for output. What is the best way to work with non-square images, such as 4:3 camera photos? I can basically think of 3 methods for accomplishing this: Scale the image to 1024x1024, ignoring aspect ratio, then inversely scale the output back to the original size. However, I have a big concern that squashing the content will result in poor inference results. Scale the image, preserving its aspect ratio so its minimum dimension is 1024, then run the model multiple times on a sliding 1024x1024 window and then aggregating the results. My main concern here is the complexity of de-duping the output, when each run could make different outputs based on how objects are cropped. Fit the image within 1024x1024 and pad with black pixels to make a square. I'm not sure if the border will muck up the inference. Anyway, this seems like it must be a well-solved problem in ML, but I'm having difficulty finding an authoritative best practice.

I Got Access and Then I opened the App and it said still waiting I was very sad

I’m trying to follow Apple’s “WWDC24: Bring your machine learning and AI models to Apple Silicon” session to convert the Mistral-7B-Instruct-v0.2 model into a Core ML package, but I’ve run into a roadblock that I can’t seem to overcome. I’ve uploaded my full conversion script here for reference: https://pastebin.com/T7Zchzfc When I run the script, it progresses through tracing and MIL conversion but then fails at the backend_mlprogram stage with this error: https://pastebin.com/fUdEzzKM The core of the error is: ValueError: Op "keyCache_tmp" (op_type: identity) Input x="keyCache" expects list, tensor, or scalar but got state[tensor[1,32,8,2048,128,fp16]] I’ve registered my KV-cache buffers in a StatefulMistralWrapper subclass of nn.Module, matching the keyCache and valueCache state names in my ct.StateType definitions, but Core ML’s backend pass reports the state tensor as an invalid input. I’m using Core ML Tools 8.3.0 on Python 3.9.6, targeting iOS18, and forcing CPU conversion (MPS wasn’t available). Any pointers on how to satisfy the handle_unused_inputs pass or properly declare/cache state for GQA models in Core ML would be greatly appreciated! Thanks in advance for your help, Usman Khan

Using highly optimized Metal Shading Language (MSL) code, I pushed the MacBook Air M2 to its performance limits with the deformable_attention_universal kernel. The results demonstrate both the efficiency of the code and the exceptional power of Apple Silicon. The total computational workload exceeded 8.455 quadrillion FLOPs, equivalent to processing 8,455 trillion operations. On average, the code sustained a throughput of 85.37 TFLOPS, showcasing the chip’s remarkable ability to handle massive workloads. Peak instantaneous performance reached approximately 673.73 TFLOPS, reflecting near-optimal utilization of the GPU cores. Despite this intensity, the cumulative GPU runtime remained under 100 seconds, highlighting the code’s efficiency and time optimization. The fastest iteration achieved a record processing time of only 0.051 ms, demonstrating minimal bottlenecks and excellent responsiveness. Memory management was equally impressive: peak GPU memory usage never exceeded 2 MB, reflecting efficient use of the M2’s Unified Memory. This minimizes data transfer overhead and ensures smooth performance across repeated workloads. Overall, these results confirm that a well-optimized Metal implementation can unlock the full potential of Apple Silicon, delivering exceptional computational density, processing speed, and memory efficiency. The MacBook Air M2, often considered an energy-efficient consumer laptop, is capable of handling highly intensive workloads at performance levels typically expected from much larger GPUs. This test validates both the robustness of the Metal code and the extraordinary capabilities of the M2 chip for high-performance computing tasks.

I would like to make use of create ML to classify a motion. However, it seems it requires 2 classes at least to train or test it. What should I do as I only has 1 class (the target motion). Also, how to interpret the 'Recall' and 'F1 Score'

I'm downloading a fine-tuned model from HuggingFace which is then cached on my Mac when the app first starts. However, I wanted to test adding a progress bar to show the download progress. To test this I need to delete the cached model. From what I've seen online this is cached at /Users/userName/.cache/huggingface/hub However, if I delete the files from here, using Terminal, the app still seems to be able to access the model. Is the model cached somewhere else? On my iPhone it seems deleting the app also deletes the cached model (app data) so that is useful.

Hi everyone, Could someone confirm if it's currently possible, or if there are any plans, to restrict users from enabling Apple Intelligence altogether? I understand that we can block individual features using MDM, but I'm interested in knowing if we can prevent users from toggling Apple Intelligence on and off in System Settings entirely. Thanks! Kind Regards, Filipe Nogueira

Hello fellow developers, I'm the founder of a FinTech startup, Cent Capital (https://cent.capital), where we are building an AI-powered financial co-pilot. We're deeply exploring the Apple ecosystem to create a more proactive and ambient user experience. A core part of our vision is to use App Intents and the Shortcuts app to surface personalized financial insights without the user always needing to open our app. For example, suggesting a Shortcut like, "What's my spending in the 'Dining Out' category this month?" or having an App Intent proactively surface an insight like, "Your 'Subscriptions' budget is almost full." My question for the community is about the architectural and user experience best practices for this. How are you thinking about the balance between providing rich, actionable insights via Intents without being overly intrusive or "spammy" to the user? What are the best practices for designing the data model that backs these App Intents for a complex domain like personal finance? Are there specific performance or privacy considerations we should be aware of when surfacing potentially sensitive financial data through these system-level integrations? We believe this is the future of FinTech apps on iOS and would love to hear how other developers are thinking about this challenge. Thanks for your insights!

I'm running MacOs 26 Beta 5. I noticed that I can no longer achieve 100% usage on the ANE as I could before with Apple Foundations on-device model. Has Apple activated some kind of throttling or power limiting of the ANE? I cannot get above 3w or 40% usage now since upgrading. I'm on the high power energy mode. I there an API rate limit being applied? I kave a M4 Pro mini with 64 GB of memory.

I am trying to run TinyLlama directly using Swift Playgrounds for iOS. I have tried multiple solutions, like libraries (LLM.swift, swift-transformers, ...) which never worked due to import issues, and also tried importing an exported mlmodel. For the later, I followed the article about Llama 3.1 on CoreML. It was hard to understand how to do the inference with it, but I was able to export a mlpackage, that I then placed in a xcode project to generate the mlmodelc (compiled model) and the model class. I had to go with the first version described in the article, without optimizations, as I got errors during model loading with the flexible input shapes. I was able to run the model for one token generation. But my biggest problem is that, though the mlmodelc is only 550 MiB, th model loads 24+GiB of memory, largely exceeding what I can have on an iOS device. Is there a way to use do LLM inferences on Swift Playgrounds at a reasonable speed (even 1 token / s would be sufficient)?

I have followed https://apple.github.io/coremltools/docs-guides/source/installing-coremltools.html but failed. Looks like the doc is too outdated.

Hi, I am modifying the sample camera app that is here: https://developer.apple.com/tutorials/sample-apps/capturingphotos-camerapreview ... In the processPreviewImages, I am using the Vision APIs to generate a segmentation mask for a person/object, then compositing that person onto a different background (with some other filtering). The filtering and compositing is done via CoreImage. At the end, I convert the CIImage to a CGImage then to a SwiftUI Image. When I run it on my iPhone, it works fine, and has not crashed. When I run it on the iPhone with the debugger, it crashes within a few seconds with: 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<>>: It had previously been working fine with the debugger, so I'm not sure what has changed. Is there a difference in how the Vision APIs are executed if the debugger is attached vs. not?