Business Software for ExecuTorch

Simple. Secure. Reliable messaging. With WhatsApp, you'll get fast, simple, secure messaging and calling for free*, available on phones all over the world. WhatsApp Business is a free to download app that built with the small business owner in mind. Create a catalog to showcase your products and services. Connect with your customers easily by using tools to automate, sort and quickly respond to messages. WhatsApp can also help medium and large businesses provide customer support and deliver important notifications to customers. Learn more about WhatsApp Business API. Some of your most personal moments are shared on WhatsApp, which is why we built end-to-end encryption into the latest versions of our app. When end-to-end encrypted, your messages and calls are secured so only you and the person you're communicating with can read or listen to them, and nobody in between, not even WhatsApp.

We put people first, and value craft and simplicity in our work. Our teams inspire creativity around the world, helping over 1 billion people create and share. Download the Instagram logo and other assets for use on your website, in your store window, or for other marketing materials. Tap into people’s passions to create deeper relationships with your business online and off. Tell your story using unique creative tools and immersive formats that inspire action. Connect to new customers and strengthen relationships with existing ones. From major brands to mom-and-pop shops, businesses around the world are driving proven results with Instagram. Stream live to connect with your followers, or sell products via live video. Watch videos from your favorite Creators and discover new content through Instagram Video and Reels. Get inspired by photos and videos from new accounts in Explore. Discover brands and small businesses, and shop products that are relevant to your personal style.

Facebook is the world's largest social network. We build technologies that help people connect with friends and family, find communities, and grow businesses. From fundraising to offering life-saving help in a Facebook post or signing up to donate blood, we’re inspired by the ways people show up for each other in times of need. The Facebook app helps you connect with friends, family and communities of people who share your interests. Connecting with your friends and family as well as discovering new ones is easy with features like Groups, Watch and Marketplace.

Muse Glimmer is a 30-billion-parameter open-weights model from Meta Superintelligence Labs, optimized for always-on local agent workflows. Small enough to run on a Mac or PC with a single consumer GPU, it is designed for local agents, function calling, coding, and LLM-as-a-judge evaluation without depending on cloud infrastructure or network access. The model combines long-horizon execution, precise tool calling, multimodal understanding, long-context memory, and instruction following. It can complete end-to-end agentic tasks, sustain multi-step reasoning across extended workflows, recover from failed or unexpected tool calls, and accept interleaved text and images through a dedicated perception encoder for interpreting screenshots, charts, and documents. Muse Glimmer works with OpenClaw and other agentic orchestration patterns, supports controllable reasoning effort, and is trained on data from more than 100 languages.

Easy to pick up, so you can create powerful applications immediately. Compatible with the Java ecosystem. Use your favorite JVM frameworks and libraries. Share application logic between web, mobile, and desktop platforms while keeping an experience native to users. Save time and get the benefit of unlimited access to features specific to these platforms. Kotlin has great support and many contributors in its fast-growing global community. Enjoy the benefits of a rich ecosystem with a wide range of community libraries. Help is never far away — consult extensive community resources or ask the Kotlin team directly. Kotlin Multiplatform Mobile is an SDK for iOS and Android app development. It offers all the combined benefits of creating cross-platform and native apps. Maintain a single codebase for networking, data storage, analytics, and the other logic of your Android and iOS apps.

Writing Swift code is interactive and fun, the syntax is concise yet expressive, and Swift includes modern features developers love. Swift code is safe by design and produces software that runs lightning-fast. Swift is the result of the latest research on programming languages, combined with decades of experience building Apple platforms. Named parameters are expressed in a clean syntax that makes APIs in Swift even easier to read and maintain. Even better, you don’t even need to type semi-colons. Inferred types make code cleaner and less prone to mistakes, while modules eliminate headers and provide namespaces. To best support international languages and emoji, Strings are Unicode-correct and use a UTF-8 based encoding to optimize performance for a wide-variety of use cases. You can even write concurrent code with simple, built-in keywords that define asynchronous behavior, making your code more readable and less error-prone.

The open-source AI model you can fine-tune, distill and deploy anywhere is now available in more versions. Choose from 1B, 3B, 11B or 90B, or continue building with Llama 3.1. Llama 3.2 is a collection of large language models (LLMs) pretrained and fine-tuned in 1B and 3B sizes that are multilingual text only, and 11B and 90B sizes that take both text and image inputs and output text. Develop highly performative and efficient applications from our latest release. Use our 1B or 3B models for on device applications such as summarizing a discussion from your phone or calling on-device tools like calendar. Use our 11B or 90B models for image use cases such as transforming an existing image into something new or getting more information from an image of your surroundings.

LLaVA (Large Language-and-Vision Assistant) is an innovative multimodal model that integrates a vision encoder with the Vicuna language model to facilitate comprehensive visual and language understanding. Through end-to-end training, LLaVA exhibits impressive chat capabilities, emulating the multimodal functionalities of models like GPT-4. Notably, LLaVA-1.5 has achieved state-of-the-art performance across 11 benchmarks, utilizing publicly available data and completing training in approximately one day on a single 8-A100 node, surpassing methods that rely on billion-scale datasets. The development of LLaVA involved the creation of a multimodal instruction-following dataset, generated using language-only GPT-4. This dataset comprises 158,000 unique language-image instruction-following samples, including conversations, detailed descriptions, and complex reasoning tasks. This data has been instrumental in training LLaVA to perform a wide array of visual and language tasks effectively.

Qwen3, the latest iteration of the Qwen family of large language models, introduces groundbreaking features that enhance performance across coding, math, and general capabilities. With models like the Qwen3-235B-A22B and Qwen3-30B-A3B, Qwen3 achieves impressive results compared to top-tier models, thanks to its hybrid thinking modes that allow users to control the balance between deep reasoning and quick responses. The platform supports 119 languages and dialects, making it an ideal choice for global applications. Its pre-training process, which uses 36 trillion tokens, enables robust performance, and advanced reinforcement learning (RL) techniques continue to refine its capabilities. Available on platforms like Hugging Face and ModelScope, Qwen3 offers a powerful tool for developers and researchers working in diverse fields.

Transition seamlessly between eager and graph modes with TorchScript, and accelerate the path to production with TorchServe. Scalable distributed training and performance optimization in research and production is enabled by the torch-distributed backend. A rich ecosystem of tools and libraries extends PyTorch and supports development in computer vision, NLP and more. PyTorch is well supported on major cloud platforms, providing frictionless development and easy scaling. Select your preferences and run the install command. Stable represents the most currently tested and supported version of PyTorch. This should be suitable for many users. Preview is available if you want the latest, not fully tested and supported, 1.10 builds that are generated nightly. Please ensure that you have met the prerequisites (e.g., numpy), depending on your package manager. Anaconda is our recommended package manager since it installs all dependencies.

Objective-C is the primary programming language you use when writing software for OS X and iOS. It’s a superset of the C programming language and provides object-oriented capabilities and a dynamic runtime. Objective-C inherits the syntax, primitive types, and flow control statements of C and adds syntax for defining classes and methods. It also adds language-level support for object graph management and object literals while providing dynamic typing and binding, deferring many responsibilities until runtime. When building apps for OS X or iOS, you’ll spend most of your time working with objects. Those objects are instances of Objective-C classes, some of which are provided for you by Cocoa or Cocoa Touch and some of which you’ll write yourself.

C++ is a simple and clear language in its expressions. It is true that a piece of code written with C++ may be seen by a stranger of programming a bit more cryptic than some other languages due to the intensive use of special characters ({}[]*&!|...), but once one knows the meaning of such characters it can be even more schematic and clear than other languages that rely more on English words. Also, the simplification of the input/output interface of C++ in comparison to C and the incorporation of the standard template library in the language, makes the communication and manipulation of data in a program written in C++ as simple as in other languages, without losing the power it offers. It is a programming model that treats programming from a perspective where each component is considered an object, with its own properties and methods, replacing or complementing structured programming paradigm, where the focus was on procedures and parameters.

Whisper is an automatic speech recognition (ASR) system developed by OpenAI for converting spoken language into text. It is trained on 680,000 hours of multilingual and multitask audio data collected from the web. The model is designed to handle diverse accents, background noise, and technical language with high accuracy. Whisper supports transcription in multiple languages as well as translation into English. It uses an encoder-decoder Transformer architecture to process audio inputs and generate text outputs. The system can also perform tasks like language identification and timestamp generation. Overall, Whisper enables developers to build robust voice-enabled applications with ease.

Phi-4-mini-reasoning is a 3.8-billion parameter transformer-based language model optimized for mathematical reasoning and step-by-step problem solving in environments with constrained computing or latency. Fine-tuned with synthetic data generated by the DeepSeek-R1 model, it balances efficiency with advanced reasoning ability. Trained on over one million diverse math problems spanning multiple levels of difficulty from middle school to Ph.D. level, Phi-4-mini-reasoning outperforms its base model on long sentence generation across various evaluations and surpasses larger models like OpenThinker-7B, Llama-3.2-3B-instruct, and DeepSeek-R1. It features a 128K-token context window and supports function calling, enabling integration with external tools and APIs. Phi-4-mini-reasoning can be quantized using Microsoft Olive or Apple MLX Framework for deployment on edge devices such as IoT, laptops, and mobile devices.

Voxtral models are frontier open source speech‑understanding systems available in two sizes—a 24 B variant for production‑scale applications and a 3 B variant for local and edge deployments, both released under the Apache 2.0 license. They combine high‑accuracy transcription with native semantic understanding, supporting long‑form context (up to 32 K tokens), built‑in Q&A and structured summarization, automatic language detection across major languages, and direct function‑calling to trigger backend workflows from voice. Retaining the text capabilities of their Mistral Small 3.1 backbone, Voxtral handles audio up to 30 minutes for transcription or 40 minutes for understanding and outperforms leading open source and proprietary models on benchmarks such as LibriSpeech, Mozilla Common Voice, and FLEURS. Accessible via download on Hugging Face, API endpoint, or private on‑premises deployment, Voxtral also offers domain‑specific fine‑tuning and advanced enterprise features.