Alternatives to Google AI Edge Gallery

LFM2 represents an advanced series of on-device foundation models designed to provide a remarkably swift generative-AI experience across a diverse array of devices. By utilizing a novel hybrid architecture, it achieves decoding and pre-filling speeds that are up to twice as fast as those of similar models, while also enhancing training efficiency by as much as three times compared to its predecessor. These models offer a perfect equilibrium of quality, latency, and memory utilization suitable for embedded system deployment, facilitating real-time, on-device AI functionality in smartphones, laptops, vehicles, wearables, and various other platforms, which results in millisecond inference, device durability, and complete data sovereignty. LFM2 is offered in three configurations featuring 0.35 billion, 0.7 billion, and 1.2 billion parameters, showcasing benchmark results that surpass similarly scaled models in areas including knowledge recall, mathematics, multilingual instruction adherence, and conversational dialogue assessments. With these capabilities, LFM2 not only enhances user experience but also sets a new standard for on-device AI performance.

Introducing Gemma 3n, our cutting-edge open multimodal model designed specifically for optimal on-device performance and efficiency. With a focus on responsive and low-footprint local inference, Gemma 3n paves the way for a new generation of intelligent applications that can be utilized on the move. It has the capability to analyze and respond to a blend of images and text, with plans to incorporate video and audio functionalities in the near future. Developers can create smart, interactive features that prioritize user privacy and function seamlessly without an internet connection. The model boasts a mobile-first architecture, significantly minimizing memory usage. Co-developed by Google's mobile hardware teams alongside industry experts, it maintains a 4B active memory footprint while also offering the flexibility to create submodels for optimizing quality and latency. Notably, Gemma 3n represents our inaugural open model built on this revolutionary shared architecture, enabling developers to start experimenting with this advanced technology today in its early preview. As technology evolves, we anticipate even more innovative applications to emerge from this robust framework.

NativeMind serves as a completely open-source AI assistant that operates directly within your browser through Ollama integration, maintaining total privacy by refraining from sending any data to external servers. All processes, including model inference and prompt handling, take place locally, which eliminates concerns about syncing, logging, or data leaks. Users can effortlessly transition between various powerful open models like DeepSeek, Qwen, Llama, Gemma, and Mistral, requiring no extra configurations, while taking advantage of native browser capabilities to enhance their workflows. Additionally, NativeMind provides efficient webpage summarization; it maintains ongoing, context-aware conversations across multiple tabs; offers local web searches that can answer questions straight from the page; and delivers immersive translations that keep the original format intact. Designed with an emphasis on both efficiency and security, this extension is fully auditable and supported by the community, ensuring enterprise-level performance suitable for real-world applications without the risk of vendor lock-in or obscure telemetry. Moreover, the user-friendly interface and seamless integration make it an appealing choice for those seeking a reliable AI assistant that prioritizes their privacy.

LiteRT, previously known as TensorFlow Lite, is an advanced runtime developed by Google that provides high-performance capabilities for artificial intelligence on devices. This platform empowers developers to implement machine learning models on multiple devices and microcontrollers with ease. Supporting models from prominent frameworks like TensorFlow, PyTorch, and JAX, LiteRT converts these models into the FlatBuffers format (.tflite) for optimal inference efficiency on devices. Among its notable features are minimal latency, improved privacy by handling data locally, smaller model and binary sizes, and effective power management. The runtime also provides SDKs in various programming languages, including Java/Kotlin, Swift, Objective-C, C++, and Python, making it easier to incorporate into a wide range of applications. To enhance performance on compatible devices, LiteRT utilizes hardware acceleration through delegates such as GPU and iOS Core ML. The upcoming LiteRT Next, which is currently in its alpha phase, promises to deliver a fresh set of APIs aimed at simplifying the process of on-device hardware acceleration, thereby pushing the boundaries of mobile AI capabilities even further. With these advancements, developers can expect more seamless integration and performance improvements in their applications.

The Gemma family consists of advanced, lightweight models developed using the same innovative research and technology as the Gemini models. These cutting-edge models are equipped with robust security features that promote responsible and trustworthy AI applications, achieved through carefully curated data sets and thorough refinements. Notably, Gemma models excel in their various sizes—2B, 7B, 9B, and 27B—often exceeding the performance of some larger open models. With the introduction of Keras 3.0, users can experience effortless integration with JAX, TensorFlow, and PyTorch, providing flexibility in framework selection based on specific tasks. Designed for peak performance and remarkable efficiency, Gemma 2 is specifically optimized for rapid inference across a range of hardware platforms. Furthermore, the Gemma family includes diverse models that cater to distinct use cases, ensuring they adapt effectively to user requirements. These lightweight language models feature a decoder and have been trained on an extensive array of textual data, programming code, and mathematical concepts, which enhances their versatility and utility in various applications.

PaliGemma 2 represents the next step forward in tunable vision-language models, enhancing the already capable Gemma 2 models by integrating visual capabilities and simplifying the process of achieving outstanding performance through fine-tuning. This advanced model enables users to see, interpret, and engage with visual data, thereby unlocking an array of innovative applications. It comes in various sizes (3B, 10B, 28B parameters) and resolutions (224px, 448px, 896px), allowing for adaptable performance across different use cases. PaliGemma 2 excels at producing rich and contextually appropriate captions for images, surpassing basic object recognition by articulating actions, emotions, and the broader narrative associated with the imagery. Our research showcases its superior capabilities in recognizing chemical formulas, interpreting music scores, performing spatial reasoning, and generating reports for chest X-rays, as elaborated in the accompanying technical documentation. Transitioning to PaliGemma 2 is straightforward for current users, ensuring a seamless upgrade experience while expanding their operational potential. The model's versatility and depth make it an invaluable tool for both researchers and practitioners in various fields.

Amazon Nova represents an advanced generation of foundation models (FMs) that offer cutting-edge intelligence and exceptional price-performance ratios, and it is exclusively accessible through Amazon Bedrock. The lineup includes three distinct models: Amazon Nova Micro, Amazon Nova Lite, and Amazon Nova Pro, each designed to process inputs in text, image, or video form and produce text-based outputs. These models cater to various operational needs, providing diverse options in terms of capability, accuracy, speed, and cost efficiency. Specifically, Amazon Nova Micro is tailored for text-only applications, ensuring the quickest response times at minimal expense. In contrast, Amazon Nova Lite serves as a budget-friendly multimodal solution that excels at swiftly handling image, video, and text inputs. On the other hand, Amazon Nova Pro boasts superior capabilities, offering an optimal blend of accuracy, speed, and cost-effectiveness suitable for an array of tasks, including video summarization, Q&A, and mathematical computations. With its exceptional performance and affordability, Amazon Nova Pro stands out as an attractive choice for nearly any application.

Reka Flash 3 is a cutting-edge multimodal AI model with 21 billion parameters, crafted by Reka AI to perform exceptionally well in tasks such as general conversation, coding, following instructions, and executing functions. This model adeptly handles and analyzes a myriad of inputs, including text, images, video, and audio, providing a versatile and compact solution for a wide range of applications. Built from the ground up, Reka Flash 3 was trained on a rich array of datasets, encompassing both publicly available and synthetic information, and it underwent a meticulous instruction tuning process with high-quality selected data to fine-tune its capabilities. The final phase of its training involved employing reinforcement learning techniques, specifically using the REINFORCE Leave One-Out (RLOO) method, which combined both model-based and rule-based rewards to significantly improve its reasoning skills. With an impressive context length of 32,000 tokens, Reka Flash 3 competes effectively with proprietary models like OpenAI's o1-mini, making it an excellent choice for applications requiring low latency or on-device processing. The model operates at full precision with a memory requirement of 39GB (fp16), although it can be efficiently reduced to just 11GB through the use of 4-bit quantization, demonstrating its adaptability for various deployment scenarios. Overall, Reka Flash 3 represents a significant advancement in multimodal AI technology, capable of meeting diverse user needs across multiple platforms.

Google AI Edge presents an extensive range of tools and frameworks aimed at simplifying the integration of artificial intelligence into mobile, web, and embedded applications. By facilitating on-device processing, it minimizes latency, supports offline capabilities, and keeps data secure and local. Its cross-platform compatibility ensures that the same AI model can operate smoothly across various embedded systems. Additionally, it boasts multi-framework support, accommodating models developed in JAX, Keras, PyTorch, and TensorFlow. Essential features include low-code APIs through MediaPipe for standard AI tasks, which enable rapid incorporation of generative AI, as well as functionalities for vision, text, and audio processing. Users can visualize their model's evolution through conversion and quantification processes, while also overlaying results to diagnose performance issues. The platform encourages exploration, debugging, and comparison of models in a visual format, allowing for easier identification of critical hotspots. Furthermore, it enables users to view both comparative and numerical performance metrics, enhancing the debugging process and improving overall model optimization. This powerful combination of features positions Google AI Edge as a pivotal resource for developers aiming to leverage AI in their applications.

QuickWhisper is a macOS tool designed for transcription, dictation, and AI summarization, utilizing the capabilities of OpenAI's Whisper model and operating completely offline without any reliance on cloud services. This versatile application can transcribe audio from various sources, including local files, YouTube videos, online meetings, and system audio, while also offering the functionality to record meetings through calendar integration, all done discreetly without disrupting screen sharing. Additionally, it provides system-wide dictation that seamlessly integrates with all macOS applications, allowing users to substitute keyboard input with voice commands, ensuring that all transcription activities are processed directly on the user's Mac. For those interested in AI summarization, QuickWhisper offers options through cloud providers like OpenAI, Anthropic, Google, xAI, Mistral, and Groq, or users can opt for on-device solutions using Ollama and LM Studio. Moreover, QuickWhisper boasts features such as batch transcription, automatic background transcription through Watch Folders, speaker diarization, integration with Apple Shortcuts, and webhooks for connecting with third-party services, making it a comprehensive tool for audio management and productivity. The combination of these features enhances the user experience, allowing for efficient and flexible handling of audio transcription and summarization tasks.

On June 23, 2025, Microsoft unveiled Mu, an innovative 330-million-parameter encoder–decoder language model specifically crafted to enhance the agent experience within Windows environments by effectively translating natural language inquiries into function calls for Settings, all processed on-device via NPUs at a remarkable speed of over 100 tokens per second while ensuring impressive accuracy. By leveraging Phi Silica optimizations, Mu’s encoder–decoder design employs a fixed-length latent representation that significantly reduces both computational demands and memory usage, achieving a 47 percent reduction in first-token latency and a decoding speed that is 4.7 times greater on Qualcomm Hexagon NPUs when compared to other decoder-only models. Additionally, the model benefits from hardware-aware tuning techniques, which include a thoughtful 2/3–1/3 split of encoder and decoder parameters, shared weights for input and output embeddings, Dual LayerNorm, rotary positional embeddings, and grouped-query attention, allowing for swift inference rates exceeding 200 tokens per second on devices such as the Surface Laptop 7, along with sub-500 ms response times for settings-related queries. This combination of features positions Mu as a groundbreaking advancement in on-device language processing capabilities.

A platform for generative media allows users to create images, audio, and videos solely from text input. You have the ability to upload your own datasets to develop personalized models tailored to your needs. Additionally, you can upload images to construct a custom model that reflects your unique style. This platform also enables the generation of videos and animations based on textual descriptions provided by the user. Users can select from various model types, including creative, structured, anime, or photorealistic styles. Notably, it features the most sophisticated algorithm for generating pixel art, setting it apart in the realm of digital creation. This versatility makes it an invaluable tool for artists and creators looking to explore new avenues in media generation.

Transform audio and video into precise text within moments using our advanced transcription service. Our GPU-accelerated engine efficiently converts various media formats, including YouTube uploads, into text almost instantly. TurboScribe utilizes Whisper, recognized as the leading AI technology for speech-to-text transcription accuracy. Additionally, users can translate their transcripts or subtitles into over 134 languages and transcribe any spoken language directly into English. Your privacy is paramount; only you can access your data, as all files and transcripts are securely encrypted. TurboScribe accommodates a wide array of popular audio and video formats such as MP3, M4A, MP4, MOV, AAC, WAV, and OGG among others. While optimal results are achieved with clear audio, TurboScribe maintains impressive accuracy even with accents, background noise, and varying audio quality. This flexibility ensures that users can rely on TurboScribe for their diverse transcription needs without concern for audio conditions.

Cognition has unveiled SWE-1.5, the newest agent-model specifically designed for software engineering, featuring an expansive "frontier-size" architecture composed of hundreds of billions of parameters and an end-to-end optimization (encompassing the model, inference engine, and agent harness) that enhances both speed and intelligence. This model showcases nearly state-of-the-art coding capabilities and establishes a new standard for latency, achieving inference speeds of up to 950 tokens per second, which is approximately six times quicker than its predecessor, Haiku 4.5, and thirteen times faster than Sonnet 4.5. Trained through extensive reinforcement learning in realistic coding-agent environments that incorporate multi-turn workflows, unit tests, and quality assessments, SWE-1.5 also leverages integrated software tools and high-performance hardware, including thousands of GB200 NVL72 chips paired with a custom hypervisor infrastructure. Furthermore, its innovative architecture allows for more effective handling of complex coding tasks and improves overall productivity for software development teams. This combination of speed, efficiency, and intelligent design positions SWE-1.5 as a game changer in the realm of coding models.

We present the Granite series of decoder-only code models specifically designed for tasks involving code generation, such as debugging, code explanation, and documentation, utilizing programming languages across a spectrum of 116 different types. An extensive assessment of the Granite Code model family across various tasks reveals that these models consistently achieve leading performance compared to other open-source code language models available today. Among the notable strengths of Granite Code models are: Versatile Code LLM: The Granite Code models deliver competitive or top-tier results across a wide array of code-related tasks, which include code generation, explanation, debugging, editing, translation, and beyond, showcasing their capacity to handle various coding challenges effectively. Additionally, their adaptability makes them suitable for both simple and complex coding scenarios. Reliable Enterprise-Grade LLM: All models in this series are developed using data that complies with licensing requirements and is gathered in alignment with IBM's AI Ethics guidelines, ensuring trustworthy usage for enterprise applications.

Mistral AI has launched two cutting-edge models designed for on-device computing and edge applications, referred to as "les Ministraux": Ministral 3B and Ministral 8B. These innovative models redefine the standards of knowledge, commonsense reasoning, function-calling, and efficiency within the sub-10B category. They are versatile enough to be utilized or customized for a wide range of applications, including managing complex workflows and developing specialized task-focused workers. Capable of handling up to 128k context length (with the current version supporting 32k on vLLM), Ministral 8B also incorporates a unique interleaved sliding-window attention mechanism to enhance both speed and memory efficiency during inference. Designed for low-latency and compute-efficient solutions, these models excel in scenarios such as offline translation, smart assistants that don't rely on internet connectivity, local data analysis, and autonomous robotics. Moreover, when paired with larger language models like Mistral Large, les Ministraux can effectively function as streamlined intermediaries, facilitating function-calling within intricate multi-step workflows, thereby expanding their applicability across various domains. This combination not only enhances performance but also broadens the scope of what can be achieved with AI in edge computing.

Private LLM is an AI chatbot designed for use on iOS and macOS that operates offline, ensuring that your data remains entirely on your device, secure, and private. Since it functions without needing internet access, your information is never transmitted externally, staying solely with you. You can enjoy its features without any subscription fees, paying once for access across all your Apple devices. This tool is created for everyone, offering user-friendly functionalities for text generation, language assistance, and much more. Private LLM incorporates advanced AI models that have been optimized with cutting-edge quantization techniques, delivering a top-notch on-device experience while safeguarding your privacy. It serves as a smart and secure platform for fostering creativity and productivity, available whenever and wherever you need it. Additionally, Private LLM provides access to a wide range of open-source LLM models, including Llama 3, Google Gemma, Microsoft Phi-2, Mixtral 8x7B family, and others, allowing seamless functionality across your iPhones, iPads, and Macs. This versatility makes it an essential tool for anyone looking to harness the power of AI efficiently.

ElevenLabs has unveiled Scribe, a cutting-edge Automatic Speech Recognition (ASR) model that aims to provide remarkably accurate transcriptions in 99 different languages. This innovative system is tailored to effectively manage a wide range of real-world audio situations, featuring capabilities such as word-level timestamps, speaker identification, and audio-event tagging. In benchmark evaluations like FLEURS and Common Voice, Scribe has outperformed leading models, including Gemini 2.0 Flash, Whisper Large V3, and Deepgram Nova-3, achieving impressive word error rates of 98.7% for Italian and 96.7% for English. Additionally, Scribe shows a significant reduction in errors for languages that have often faced challenges, such as Serbian, Cantonese, and Malayalam, where competing models frequently report error rates above 40%. Furthermore, developers can easily incorporate Scribe into their applications via ElevenLabs' speech-to-text API, which returns structured JSON transcripts enriched with comprehensive annotations. This level of accessibility and performance is set to revolutionize the field of transcription and enhance the user experience across various applications.

SmolLM2 comprises an advanced suite of compact language models specifically created for on-device functionalities. This collection features models with varying sizes, including those with 1.7 billion parameters, as well as more streamlined versions at 360 million and 135 million parameters, ensuring efficient performance on even the most limited hardware. They excel in generating text and are fine-tuned for applications requiring real-time responsiveness and minimal latency, delivering high-quality outcomes across a multitude of scenarios such as content generation, coding support, and natural language understanding. The versatility of SmolLM2 positions it as an ideal option for developers aiming to incorporate robust AI capabilities into mobile devices, edge computing solutions, and other settings where resources are constrained. Its design reflects a commitment to balancing performance and accessibility, making cutting-edge AI technology more widely available.

txtai is a comprehensive open-source embeddings database that facilitates semantic search, orchestrates large language models, and streamlines language model workflows. It integrates sparse and dense vector indexes, graph networks, and relational databases, creating a solid infrastructure for vector search while serving as a valuable knowledge base for applications involving LLMs. Users can leverage txtai to design autonomous agents, execute retrieval-augmented generation strategies, and create multi-modal workflows. Among its standout features are support for vector search via SQL, integration with object storage, capabilities for topic modeling, graph analysis, and the ability to index multiple modalities. It enables the generation of embeddings from a diverse range of data types including text, documents, audio, images, and video. Furthermore, txtai provides pipelines driven by language models to manage various tasks like LLM prompting, question-answering, labeling, transcription, translation, and summarization, thereby enhancing the efficiency of these processes. This innovative platform not only simplifies complex workflows but also empowers developers to harness the full potential of AI technologies.

Anyimg.ai is a robust online platform for AI-driven image creation that transforms your imaginative ideas into high-quality art within moments. The platform features an array of AI models and a wide range of artistic styles, such as minimalist, watercolor, vintage, abstract, and realistic anime, enabling users to align visuals with their artistic preferences. With a user-friendly and straightforward interface, individuals can easily enter a detailed prompt, choose their desired models and style settings, and swiftly produce one-of-a-kind images. Additionally, it offers a gallery that displays exceptional AI-generated works, enhancing inspiration and creativity. This platform is widely respected among artists, designers, and marketers alike for its ability to deliver outstanding results that enhance and expedite creative processes, making it an invaluable tool for professionals in various fields.

Mistral AI has unveiled two cutting-edge models specifically designed for on-device computing and edge use cases, collectively referred to as "les Ministraux": Ministral 3B and Ministral 8B. These innovative models stand out due to their capabilities in knowledge retention, commonsense reasoning, function-calling, and overall efficiency, all while remaining within the sub-10B parameter range. They boast support for a context length of up to 128k, making them suitable for a diverse range of applications such as on-device translation, offline smart assistants, local analytics, and autonomous robotics. Notably, Ministral 8B incorporates an interleaved sliding-window attention mechanism, which enhances both the speed and memory efficiency of inference processes. Both models are adept at serving as intermediaries in complex multi-step workflows, skillfully managing functions like input parsing, task routing, and API interactions based on user intent, all while minimizing latency and operational costs. Benchmark results reveal that les Ministraux consistently exceed the performance of similar models across a variety of tasks, solidifying their position in the market. As of October 16, 2024, these models are now available for developers and businesses, with Ministral 8B being offered at a competitive rate of $0.1 for every million tokens utilized. This pricing structure enhances accessibility for users looking to integrate advanced AI capabilities into their solutions.

SAM Audio represents a cutting-edge advancement in AI technology aimed at precise audio segmentation and editing. This innovative tool empowers users to separate individual sounds from intricate audio compositions by utilizing intuitive prompts that reflect natural thought processes regarding sound. Users can easily input descriptive phrases like “eliminate dog barking” or “retain only the vocals,” interact with objects in a video to extract their corresponding audio, or highlight specific time intervals where desired sounds are present, all within a cohesive platform. Accessible through Meta’s Segment Anything Playground, SAM Audio allows users to upload their own audio or video files to immediately explore its features. Additionally, it can be downloaded for implementation in personalized audio projects and research endeavors. Unlike conventional audio editing tools that are limited to specific tasks, SAM Audio excels in accommodating a variety of prompts and accurately handling diverse real-world soundscapes, making it a versatile choice for audio manipulation. This level of flexibility and user-friendliness sets it apart from traditional solutions in the industry.

Google has unveiled enhanced Gemini audio models that greatly broaden the platform's functionalities for engaging and nuanced voice interactions, as well as real-time conversational AI, highlighted by the arrival of Gemini 2.5 Flash Native Audio and advancements in text-to-speech technology. The revamped native audio model supports live voice agents capable of managing intricate workflows, reliably adhering to detailed user directives, and facilitating smoother multi-turn dialogues by improving context retention from earlier exchanges. This upgrade is now accessible through Google AI Studio, Vertex AI, Gemini Live, and Search Live, allowing developers and products to create dynamic voice experiences such as smart assistants and corporate voice agents. Additionally, Google has refined the core Text-to-Speech (TTS) models within the Gemini 2.5 lineup to enhance expressiveness, tone modulation, pacing adjustments, and multilingual capabilities, resulting in synthesized speech that sounds increasingly natural. Furthermore, these innovations position Google's audio technology as a leader in the realm of conversational AI, driving forward the potential for more intuitive human-computer interactions.

Image recognition technology that gives you a complete understanding of your digital media. Our on-device computer vision system can provide a response time of less that 250ms. This is 4x faster than our API and doesn't require an internet connection. By simply scanning their phones around a room, users can identify objects in that space. This feature is exclusive to our on-device platform. Privacy concerns are almost eliminated by removing the requirement for data to be sent from the end-user device. Our API takes every precaution to protect your privacy. However, our on-device model raises security standards significantly. CloudSight will send you visual content. Our API will then generate a natural language description. Filter and categorize images. You can also monitor for inappropriate content and assign labels to all your digital media.

AccurateScribe.ai is an advanced cloud-based speech-to-text transcription platform designed to provide fast, highly accurate multilingual transcription services across more than 130 languages and dialects. Leveraging state-of-the-art AI models such as Whisper, it converts audio and video files into precise, readable text with ease and security. The platform accepts a wide range of file formats including MP3, WAV, MP4, and MOV, supporting files as large as 10 hours or 5 GB. Users can also record audio directly through an in-browser voice recorder, which transcribes content in real time, perfect for meetings, lectures, or personal notes. Additionally, AccurateScribe.ai enables transcription from public URLs on platforms like YouTube, Dropbox, and Google Drive without the need for manual file downloads. Its cloud infrastructure ensures fast processing times and secure data handling. The platform caters to a diverse range of transcription needs, from professional and academic to personal use. AccurateScribe.ai simplifies voice-to-text conversion while ensuring flexibility and reliability.

CodeGemma represents an impressive suite of efficient and versatile models capable of tackling numerous coding challenges, including middle code completion, code generation, natural language processing, mathematical reasoning, and following instructions. It features three distinct model types: a 7B pre-trained version designed for code completion and generation based on existing code snippets, a 7B variant fine-tuned for translating natural language queries into code and adhering to instructions, and an advanced 2B pre-trained model that offers code completion speeds up to twice as fast. Whether you're completing lines, developing functions, or crafting entire segments of code, CodeGemma supports your efforts, whether you're working in a local environment or leveraging Google Cloud capabilities. With training on an extensive dataset comprising 500 billion tokens predominantly in English, sourced from web content, mathematics, and programming languages, CodeGemma not only enhances the syntactical accuracy of generated code but also ensures its semantic relevance, thereby minimizing mistakes and streamlining the debugging process. This powerful tool continues to evolve, making coding more accessible and efficient for developers everywhere.

Gemma represents a collection of cutting-edge, lightweight open models that are built upon the same research and technology underlying the Gemini models. Created by Google DeepMind alongside various teams at Google, the inspiration for Gemma comes from the Latin word "gemma," which translates to "precious stone." In addition to providing our model weights, we are also offering tools aimed at promoting developer creativity, encouraging collaboration, and ensuring the ethical application of Gemma models. Sharing key technical and infrastructural elements with Gemini, which stands as our most advanced AI model currently accessible, Gemma 2B and 7B excel in performance within their weight categories when compared to other open models. Furthermore, these models can conveniently operate on a developer's laptop or desktop, demonstrating their versatility. Impressively, Gemma not only outperforms significantly larger models on crucial benchmarks but also maintains our strict criteria for delivering safe and responsible outputs, making it a valuable asset for developers.

Push the limits of individual alignment, artificial consciousness, open-source software, and decentralization through experimentation that larger corporations and governments often shy away from. Hermes 3 features sophisticated long-term context retention, the ability to engage in multi-turn conversations, and intricate roleplaying and internal monologue capabilities, alongside improved functionality for agentic function-calling. The design of this model emphasizes precise adherence to system prompts and instruction sets in a flexible way. By fine-tuning Llama 3.1 across various scales, including 8B, 70B, and 405B, and utilizing a dataset largely composed of synthetically generated inputs, Hermes 3 showcases performance that rivals and even surpasses Llama 3.1, while also unlocking greater potential in reasoning and creative tasks. This series of instructive and tool-utilizing models exhibits exceptional reasoning and imaginative skills, paving the way for innovative applications. Ultimately, Hermes 3 represents a significant advancement in the landscape of AI development.

Opal is an innovative mini-app creator developed by Google Labs that allows users to build, modify, and share simple applications solely through natural language input and a visual interface. Currently in a public beta phase limited to the U.S., Opal enhances the prototyping process by converting straightforward English descriptions of application logic into a series of connected actions, which include prompts, calls to AI models, and integration of external tools, all represented in a flowchart that can be edited. Users can easily adjust any part of the process or introduce new features by using conversational commands or by directly interacting with the visual editor, eliminating the need for coding skills. Additionally, Opal includes a gallery of pre-made demo templates for users to adapt to their specific needs, and once an application is finalized, it can be published immediately with a shareable link for anyone with a Google account. This tool is designed to streamline the transition from concepts to practical AI-driven applications, making it accessible to a broader audience. With its user-friendly approach, Opal empowers individuals to innovate and create without technical barriers.

Say goodbye to tedious manual transcription; with Audioscribe, you can effortlessly transcribe, search, and comprehend your audio. Transform your spoken words into valuable insights using our cutting-edge transcription service. AudioScribe.io is an innovative solution that breathes life into your dialogues. Designed to cater to everyone, from independent professionals to large corporations, AudioScribe.io guarantees that no important detail gets overlooked in meetings, interviews, or critical discussions. Our advanced AI technology offers the highest-quality transcription service available today. When compared to competitors like Zoom transcription, AudioScribe.io stands out due to its unmatched precision. Additionally, AudioScribe.io harnesses the power of a Large Language Model (LLM) that allows you to delve into your text thoroughly. Simply pose questions related to your transcript, and our AI will deliver insights derived directly from your content, enhancing your understanding. Explore your conversations more deeply, analyze sentiments, identify key themes, and much more to unlock the full potential of your discussions. With AudioScribe.io, every word matters, and now you can make the most of them.

Apollo is a streamlined mobile application that facilitates completely on-device, cloud-independent AI interactions, allowing users to interact with sophisticated language and vision models in a secure, private manner with minimal delays. It features a collection of compact foundation models sourced from the company's LEAP platform, enabling users to compose messages, send emails, converse with a personal AI assistant, create digital characters, or utilize image-to-text functions, all while maintaining offline capabilities and ensuring no data is transmitted beyond the device. Optimized for immediate responsiveness and offline functionality, Apollo guarantees that all inference occurs locally, eliminating the need for API calls, external servers, or logging of user data. This application acts as both a personal AI exploration tool and a development environment for those utilizing LEAP models, allowing users to effectively assess a model's performance on their specific mobile devices prior to more widespread implementation. Additionally, Apollo's design emphasizes user autonomy, ensuring a seamless experience free from external interruptions or privacy concerns.

UniScribe, powered by AI, is a platform which helps users extract key information quickly from long audio and video files on their local computer or YouTube videos. Features: - Conversion of YouTube videos or local audio files to text is faster using an optimized Whisper model. - Automatic generation and distribution of mind maps, key Q&A, and summaries. - Supports exporting text content in various formats, such as .txt/.pdf/.docx/.srt/.vtt/.csv. Use Cases - Journalists & Writers: Transcribing interview recordings to text for easier quoting & editing. Students and Academics - To transcribe lectures or seminars for easier note-taking. - Market Researchers: Transcribing audio data from focus group and interview sessions for analysis. - Legal Professionals : Transcribe court records, testimony, and client interviews to prepare legal documents and conduct research. -Content Producers and Creators: To transcribing media content for blog postings

Qwen3-Coder is a versatile coding model that comes in various sizes, prominently featuring the 480B-parameter Mixture-of-Experts version with 35B active parameters, which naturally accommodates 256K-token contexts that can be extended to 1M tokens. This model achieves impressive performance that rivals Claude Sonnet 4, having undergone pre-training on 7.5 trillion tokens, with 70% of that being code, and utilizing synthetic data refined through Qwen2.5-Coder to enhance both coding skills and overall capabilities. Furthermore, the model benefits from post-training techniques that leverage extensive, execution-guided reinforcement learning, which facilitates the generation of diverse test cases across 20,000 parallel environments, thereby excelling in multi-turn software engineering tasks such as SWE-Bench Verified without needing test-time scaling. In addition to the model itself, the open-source Qwen Code CLI, derived from Gemini Code, empowers users to deploy Qwen3-Coder in dynamic workflows with tailored prompts and function calling protocols, while also offering smooth integration with Node.js, OpenAI SDKs, and environment variables. This comprehensive ecosystem supports developers in optimizing their coding projects effectively and efficiently.

Qwen3-Omni is a comprehensive multilingual omni-modal foundation model designed to handle text, images, audio, and video, providing real-time streaming responses in both textual and natural spoken formats. Utilizing a unique Thinker-Talker architecture along with a Mixture-of-Experts (MoE) framework, it employs early text-centric pretraining and mixed multimodal training, ensuring high-quality performance across all formats without compromising on text or image fidelity. This model is capable of supporting 119 different text languages, 19 languages for speech input, and 10 languages for speech output. Demonstrating exceptional capabilities, it achieves state-of-the-art performance across 36 benchmarks related to audio and audio-visual tasks, securing open-source SOTA on 32 benchmarks and overall SOTA on 22, thereby rivaling or equaling prominent closed-source models like Gemini-2.5 Pro and GPT-4o. To enhance efficiency and reduce latency in audio and video streaming, the Talker component leverages a multi-codebook strategy to predict discrete speech codecs, effectively replacing more cumbersome diffusion methods. Additionally, this innovative model stands out for its versatility and adaptability across a wide array of applications.

EmbeddingGemma is a versatile multilingual text embedding model with 308 million parameters, designed to be lightweight yet effective, allowing it to operate seamlessly on common devices like smartphones, laptops, and tablets. This model, based on the Gemma 3 architecture, is capable of supporting more than 100 languages and can handle up to 2,000 input tokens, utilizing Matryoshka Representation Learning (MRL) for customizable embedding sizes of 768, 512, 256, or 128 dimensions, which balances speed, storage, and accuracy. With its GPU and EdgeTPU-accelerated capabilities, it can generate embeddings in a matter of milliseconds—taking under 15 ms for 256 tokens on EdgeTPU—while its quantization-aware training ensures that memory usage remains below 200 MB without sacrificing quality. Such characteristics make it especially suitable for immediate, on-device applications, including semantic search, retrieval-augmented generation (RAG), classification, clustering, and similarity detection. Whether used for personal file searches, mobile chatbot functionality, or specialized applications, its design prioritizes user privacy and efficiency. Consequently, EmbeddingGemma stands out as an optimal solution for a variety of real-time text processing needs.

Ai2 OLMoE is a completely open-source mixture-of-experts language model that operates entirely on-device, ensuring that you can experiment with the model in a private and secure manner. This application is designed to assist researchers in advancing on-device intelligence and to allow developers to efficiently prototype innovative AI solutions without the need for cloud connectivity. OLMoE serves as a highly efficient variant within the Ai2 OLMo model family. Discover the capabilities of state-of-the-art local models in performing real-world tasks, investigate methods to enhance smaller AI models, and conduct local tests of your own models utilizing our open-source codebase. Furthermore, you can seamlessly integrate OLMoE into various iOS applications, as the app prioritizes user privacy and security by functioning entirely on-device. Users can also easily share the outcomes of their interactions with friends or colleagues. Importantly, both the OLMoE model and the application code are fully open source, offering a transparent and collaborative approach to AI development. By leveraging this model, developers can contribute to the growing field of on-device AI while maintaining high standards of user privacy.

Nebius Token Factory is an advanced AI inference platform that enables the production of both open-source and proprietary AI models without the need for manual infrastructure oversight. It provides enterprise-level inference endpoints that ensure consistent performance, automatic scaling of throughput, and quick response times, even when faced with high request traffic. With a remarkable 99.9% uptime, it accommodates both unlimited and customized traffic patterns according to specific workload requirements, facilitating a seamless shift from testing to worldwide implementation. Supporting a diverse array of open-source models, including Llama, Qwen, DeepSeek, GPT-OSS, Flux, and many more, Nebius Token Factory allows teams to host and refine models via an intuitive API or dashboard interface. Users have the flexibility to upload LoRA adapters or fully fine-tuned versions directly, while still benefiting from the same enterprise-grade performance assurances for their custom models. This level of support ensures that organizations can confidently leverage AI technology to meet their evolving needs.

The latest iteration of the open-source AI model, which can be fine-tuned and deployed in various environments, is now offered in multiple versions, including 1B, 3B, 11B, and 90B, alongside the option to continue utilizing Llama 3.1. Llama 3.2 comprises a series of large language models (LLMs) that come pretrained and fine-tuned in 1B and 3B configurations for multilingual text only, while the 11B and 90B models accommodate both text and image inputs, producing text outputs. With this new release, you can create highly effective and efficient applications tailored to your needs. For on-device applications, such as summarizing phone discussions or accessing calendar tools, the 1B or 3B models are ideal choices. Meanwhile, the 11B or 90B models excel in image-related tasks, enabling you to transform existing images or extract additional information from images of your environment. Overall, this diverse range of models allows developers to explore innovative use cases across various domains.

Qwen3-Coder is an advanced code model that comes in various sizes, prominently featuring the 480B-parameter Mixture-of-Experts version (with 35B active) that inherently accommodates 256K-token contexts, which can be extended to 1M, and demonstrates cutting-edge performance in Agentic Coding, Browser-Use, and Tool-Use activities, rivaling Claude Sonnet 4. With a pre-training phase utilizing 7.5 trillion tokens (70% of which are code) and synthetic data refined through Qwen2.5-Coder, it enhances both coding skills and general capabilities, while its post-training phase leverages extensive execution-driven reinforcement learning across 20,000 parallel environments to excel in multi-turn software engineering challenges like SWE-Bench Verified without the need for test-time scaling. Additionally, the open-source Qwen Code CLI, derived from Gemini Code, allows for the deployment of Qwen3-Coder in agentic workflows through tailored prompts and function calling protocols, facilitating smooth integration with platforms such as Node.js and OpenAI SDKs. This combination of robust features and flexible accessibility positions Qwen3-Coder as an essential tool for developers seeking to optimize their coding tasks and workflows.

Pulze.ai is an enterprise-level AI platform that eliminates the need for coding, empowering teams to create, implement, and oversee AI-driven assistants and workflows seamlessly. By centralizing access to over 50 top-tier LLMs and AI models, it employs intelligent routing to ensure that every request is processed by the most appropriate model, thereby maximizing performance, quality, and cost-effectiveness. Users can interact with AI models, upload and use documents, conduct web searches, generate images, transcribe audio, and automate various tasks—all from a cohesive “Space” workspace. Additionally, Pulze enables the creation of no-code AI assistants using templates tailored for specific functions, such as managing support tickets, drafting sales materials, and sorting leads. It seamlessly integrates with popular tools like Slack, Google Drive, and Jira, while also upholding enterprise-level security measures, including SOC 2 compliance, data isolation, and a policy of zero data logging by AI providers. This comprehensive approach ensures that organizations can harness the power of AI while maintaining strict security and operational standards.

Gladia is a sophisticated audio transcription and intelligence solution that provides a cohesive API, accommodating both asynchronous (for pre-recorded content) and live streaming transcription, thereby allowing developers to translate spoken words into text across more than 100 languages. This platform boasts features such as word-level timestamps, language recognition, code-switching capabilities, speaker identification, translation, summarization, a customizable vocabulary, and entity extraction. With its real-time engine, Gladia maintains latencies below 300 milliseconds while ensuring a high level of accuracy, and it offers “partials” or intermediate transcripts to enhance responsiveness during live events. Additionally, the asynchronous API is driven by a proprietary Whisper-Zero model tailored for enterprise audio applications, enabling clients to utilize add-ons like improved punctuation, consistent naming conventions, custom metadata tagging, and the ability to export to various subtitle formats such as SRT and VTT. Overall, Gladia stands out as a versatile tool for developers looking to integrate comprehensive audio transcription capabilities into their applications.

Gemma 3, launched by Google, represents a cutting-edge AI model constructed upon the Gemini 2.0 framework, aimed at delivering superior efficiency and adaptability. This innovative model can operate seamlessly on a single GPU or TPU, which opens up opportunities for a diverse group of developers and researchers. Focusing on enhancing natural language comprehension, generation, and other AI-related functions, Gemma 3 is designed to elevate the capabilities of AI systems. With its scalable and robust features, Gemma 3 aspires to propel the evolution of AI applications in numerous sectors and scenarios, potentially transforming the landscape of technology as we know it.

Meta's MusicGen is an open-source deep-learning model designed to create short musical compositions based on textual descriptions. Trained on 20,000 hours of music, encompassing complete tracks and single instrument samples, this model produces 12 seconds of audio in response to user prompts. Additionally, users can submit reference audio to extract a general melody, which the model will incorporate alongside the provided description. All generated samples utilize the melody model, ensuring consistency. Furthermore, users have the option to run the model on their own GPUs or utilize Google Colab by following the guidelines available in the repository. MusicGen features a single-stage transformer architecture combined with efficient token interleaving techniques, which streamline the process by eliminating the need for multiple cascading models. This innovative approach enables MusicGen to generate high-quality audio samples that are responsive to both textual inputs and musical characteristics, allowing users to exert greater control over the final output. The combination of these features positions MusicGen as a versatile tool for music creation and exploration.

Mistral Small 3.1 represents a cutting-edge, multimodal, and multilingual AI model that has been released under the Apache 2.0 license. This upgraded version builds on Mistral Small 3, featuring enhanced text capabilities and superior multimodal comprehension, while also accommodating an extended context window of up to 128,000 tokens. It demonstrates superior performance compared to similar models such as Gemma 3 and GPT-4o Mini, achieving impressive inference speeds of 150 tokens per second. Tailored for adaptability, Mistral Small 3.1 shines in a variety of applications, including instruction following, conversational support, image analysis, and function execution, making it ideal for both business and consumer AI needs. The model's streamlined architecture enables it to operate efficiently on hardware such as a single RTX 4090 or a Mac equipped with 32GB of RAM, thus supporting on-device implementations. Users can download it from Hugging Face and access it through Mistral AI's developer playground, while it is also integrated into platforms like Google Cloud Vertex AI, with additional accessibility on NVIDIA NIM and more. This flexibility ensures that developers can leverage its capabilities across diverse environments and applications.