Pinokio: What’s New for 2025 in the Ultimate AI App Browser

Major platform upgrades and exciting new apps put the frontier of generative AI at your fingertips

View fullsize

One of our most popular posts last year was "Get to Know Pinokio", which introduced the Pinokio desktop app that lets you install, manage, and run a large library of the latest AI applications with just a click. Pinokio has rapidly evolved into a Swiss Army Knife for running the latest open source AI applications. Since our last post, the Pinokio team has rolled out major updates and added an array of exciting new apps for AI video generation, image editing, voice cloning, and more. Here's a round-up of some of the most important changes and new apps to help you get up to speed on what's new in Pinokio for 2025!

Major Pinokio AI Platform Upgrades

Released in late 2024, Pinokio 3.0 provided a major update brought a host of enhancements and polish to the user experience. Pinokio's core software saw big upgrades leading into 2025, making it more powerful and user-friendly than ever:

Customize your Pinokio Browser - Image Credit: Pinokio

Pinokio 3.0: Released in late 2024, Pinokio 3.0 introduced a host of new features. The web UI is now "100% hackable," meaning tech-savvy users can fully customize the interface. It added Native Universal Virtualization (UV) for faster installation of dependencies, a Disk Usage Display for monitoring storage, a Debug Mode to pinpoint installation errors, and built-in Hugging Face API support.

Smoother Installs & Reliability Fixes: The Pinokio team focused heavily on reliability in subsequent updates. By version 3.30, they had fixed most of the pesky setup issues that could leave new users stuck. To make it easy to manage AI apps, Pinokio orchestrates all the complex dependencies between various system requirements (shell, conda, pip, etc.), and 3.3.0 tackled long-standing stability problems in that orchestration.

Cross-Platform Polish: Pinokio 3.x updates benefited all users on Windows, Mac, and Linux. For instance, the Pinokio team bundled a workaround for Apple's app notarization requirement, making it easier for Mac users to install Pinokio. Meanwhile, long-standing issues with Windows Visual Studio/CUDA errors when building certain AI apps were also fixed, providing a big improvement in reliability for PC users. Installing and updating Pinokio is now much more predictable and reliable than before.

Final Touches in 3.6: In February 2025,  version 3.6 arrived as the final major v3 release, strongly recommended for all users. It squashed remaining edge-case bugs and further hardens the platform to keep apps running even in odd scenarios. Pinokio 3.6 marked a stable plateau with most serious issues fixed, so you can focus on using the apps rather than troubleshooting. All told, Pinokio AI in 2025 is a much more robust and reliable platform, so be sure to update to the latest version to enjoy all the goodies.

Now, with the platform running smoothly, let's dive into the exciting new AI apps Pinokio has added since our last update!

Next-Level AI Video Generation (Text-to-Video and Beyond)

One of the most buzzworthy additions to Pinokio AI is the addition of powerful new video generation tools. If you've ever wanted to create AI-generated videos on your PC, Pinokio's new one-click apps have you covered:

CogStudio: A Full AI Video Suite: CogStudio launched in Pinokio late 2024. It's a 100% open-source video creation suite powered by advanced diffusion models. CogStudio offers a complete AI video generation suite with:

• Text-to-video generation (enter a prompt, get a video)

• Image-to-video (give an image and prompt, it animates the image into a video)

• Video-to-video (transform an input video using a prompt)

• Video extend (generate a continuation of an existing video clip)

• A seamless workflow tying these together, so you can mix and match.

CogStudio UI. Image Credit: CogVideo

What really sets CogStudio apart is its infinite video feature. You can generate a video, then pick a frame and prompt the AI to continue the story and keep chaining clips endlessly. For example, start with a teddy bear walking, then prompt it to jump at the end of the clip, then to dance, and so on—stitching together each new segment. This means infinitely long, controllable videos are now possible on consumer GPUs. Even better, CogStudio is optimized for lower VRAM (as little as ~5GB for image-to-video). If you have an NVIDIA GPU, definitely give CogVideo a try.

Hunyuan Video: In December 2024, Pinokio added one-click support for Hunyuan Video, with the one-click Gradio-based app HunyuanVideoGP. With HunyuanVideoGP in Pinokio, you can launch the app, enter a prompt, and start to generate videos. This app is NVIDIA-only and needs a decent setup (≥8GB VRAM and ~30GB RAM), but it's fast and supports advanced features like LoRAs for video fine-tuning. If you're into AI video, this is a must-try on Pinokio.

Allegro Gradio UI

Allegro & WAN Models: Pinokio has also integrated other experimental video generators. The Allegro text-to-video model got a one-click launcher in late 2024. And in Feb 2025, Pinokio added support for WAN (a new video model from Alibaba). Through a ComfyUI "mix" script, you can one-click download all the necessary workflows/models for WAN and generate videos effortlessly.

3D Content Generation: For 3D creators, Hunyuan 3D V2 was added to Pinokio in January 2025. Hunyuan 3D V2 is an AI model that generates 3D objects from a single image. Pinokio offers two launchers: one for the official high-quality model, and one "low VRAM" variant that needs only 6GB VRAM yet looks almost as good. Note, this is NVIDIA-only (current 3D AI models don't support Mac GPUs yet). If you update Pinokio to v3.6 you can dive into 3D modeling and try both model variants side by side.

Hunyuan 3D. Image Credit: Tencent Hunyuan

Diamond world simulator. Image Credit: Simulated Worlds

Simulated Worlds: On a fun note, Diamond is now available with a one-click launcher for its "world simulator" AI. Diamond runs AI models trained on video games like Counter-Strike and classic Atari games to simulate those game environments. It works on all platforms. While a bit niche, it's a cool example of the diversity of Pinokio's app ecosystem. With Diamond, you can watch an AI play Pong or Pac-Man in a simulated setting on your laptop!

Whether you want to create short films or play with AI game worlds, the new video generation apps available in Pinokio open up a lot of possibilities without the need for high-powered servers or technical knowledge.

Simplified Image Editing and Creation

Pinokio AI hasn't forgotten about images. For 2025, they’ve added several powerful yet easy to use desktop image generation tools:

Invoke AI 5.0. Image Credit: Invoke AI

Invoke AI

Invoke AI 5.0, a popular open-source image generation/editing interface is now available in Pinokio. The 5.0 release adds support for Flux.1 and Flux LoRAs, a layers system for treating AI outputs like Photoshop layers, drag-and-drop asset creation, and much more. With one click installation in Pinokio, you can deploy Invoke 5.0 and get a powerful web UI to generate and edit images locally. (Mac users note: Flux model support in Invoke is still Windows/Linux only for now.) If you're into image generation, it's never been easier to iterate on AI art with fine control.

FaceFusion

FaceFusion App

The popular FaceFusion app (think deepfakes/face swapping) is now available in Pinokio on all platforms. New features include an Age Modifier to make a face look younger/older, a Face Editor powered by LivePortrait for live facial animations, and even a Workflow system to batch process or chain effects. FaceFusion lets you do Hollywood-like face replacements and edits with surprising ease. It is 100% open source and runs via a Gradio interface. Ever wanted to swap your face into a movie scene or give yourself a virtual makeover? Give FaceFusion on Pinokio a go!

Magic Quill

Inpainting with the MagicQuill Gradio UI

MagicQuill is an AI-powered image editor where you load an image and tell the AI in plain language what to do like "make the sky sunset orange" or "add a pair of glasses." No complicated settings—it just works. The beauty of MagicQuill is that there's literally zero learning curve, making advanced image edits accessible to anyone. MagicQuill runs on all operating systems and uses Gradio. This is a perfect example of a great UX making AI useful: you can iteratively refine an image through dialogue. It's like having an AI art director at your service.

With all the new apps available for image and video, Pinokio now provides an end-to-end image and video creation and editing suite. All easy to install and use, it's like having your own creative team powered by AI, right on your desktop.

Voice Cloning and Speech AI

Pinokio has also added a number of new apps representing major advances in voice and speech AI. If you’re ready to experiment with voice cloning, text-to-speech (TTS), or audio creativity, there's a lot to get excited about:

Applio: Speech-to-Speech Conversion

Applio UI

Pinokio now offers the Applio Speech-to-Speech conversion tool as a one-click app. Put simply, Applio lets you convert one person's voice into another's. The app provides access to tens of thousands of voice models (all open source). It's truly plug-and-play: record or upload a clip, pick a target voice model, and get the converted output. If you've ever wanted to clone a voice or do parody dubs, Applio is incredibly fun and useful!

In addition to Applio, Pinokio now also features a range of cutting-edge TTS and audio processing apps. Zonos offers jaw-dropping voice cloning and nuanced emotional inflections, while F5-TTS includes a handy finetune interface for training the AI to speak your language or accent. Fish TTS can create multilingual voices right out of the box with minimal VRAM overhead, and StyleTTS Studio lets you switch between tones—cheerful, solemn, or anything in between—without fussing over voice samples. Finally, if you're in the mood for a quick laugh, Hallucinator “auto-completes" audio clips by continuing the speaker's voice in hilarious ways.

Put together, these apps make Pinokio a one-stop shop for generative speech that’s ready to tackle projects in any language or style you need.

StyleTTS Studio

Another cool tool is StyleTTS Studio, a text-to-speech engine that generates speech in various styles. It doesn’t clone voices per se, but lets you create a style and even save it for reuse. If you need a quick text-to-speech with emotion or a certain vibe, like narrating in a cheerful tone vs. a spooky tone, StyleTTS can deliver. StyleTTS runs on both PCs and Macs. It’s great for projects where you need AI voices but not necessarily a specific real voice cloned.

Music Generation on Your PC

Ever wanted an AI to compose a song for you? Thanks to new music apps like YuE in Pinokio, you can be your own music producer:

YuE: Text-to-Song Generation

YuE Text-to-Song Generation

While text-to-speech is already impressive, YuE takes audio AI to the next level: it composes entire songs with vocals, melody, and accompaniment from your lyrics. If you have a reference track or hum a melody, YuE can incorporate that input to shape the style and chord progression.

You can make a K-pop style track generated using your own lyrics or experiment and combine a Beatles vibe with modern EDM. Though the AI vocals can sometimes sound "robotic," improvements and updates are frequent. YuE is a remarkable step toward fully AI-driven music production (no bandmates required).

AI Coding Assistants and Autonomous Agents

New apps for AI-assisted coding and autonomous agents have also been added to Pinokio recently, bringing some futuristic productivity tools to everyday users:

Bolt.diy AI Coding Assistant: Bolt.diy is an AI coding assistant designed to let you build full‑stack web applications directly in your browser. It’s the community version of Bolt.new that gives you the flexibility to choose from a range of language models like OpenAI, Anthropic, Ollama, Gemini, LMStudio, and more. Bolt.diy not only generates code but can also manage packages, run backend servers, interact with APIs, and deploys applications—all within a unified, browser-based environment.

WebUI: Autonomous Web Surfing: Pinokio 3.x introduced a native integration for Microsoft’s Playwright, an open source framework for web testing and browser automation. This enabled a wild new app now in Pinokio called WebUI. WebUI is similar to OpenAI’s Operator, providing an autonomous agent that can browse the web for you.

You give it a task in natural language like, “Find me the top 3 budget smartphones with their prices,” and it will launch a browser, search the web, click links, and gather information to complete the task. It’s like your own private version of Operator, but installable with one click using Pinokio. While this is still experimental and agents can be hit-or-miss, seeing it work feels like magic.

WebUI - Mapping out a webpage

Autopilot for macOS

If you want to see your computer truly "drive itself," macOS-use is borderline sci-fi. It physically clicks around your Mac's UI, responding to your instructions. You can ask it to do things like "Create a new calendar event for tomorrow at 10 AM." or "Open my Mail app and draft an email to John, subject 'Project Update.'" The app uses a local controller with an LLM guiding it. It's recommended to use it with Claude or GPT-4 for best accuracy, but you could try other tool-calling models. While it's not perfect, it hints at a near-future where your Mac might handle routine tasks through simple prompts.

Conclusion: The Frontier of AI at Your Fingertips

Going into 2025, Pinokio stands out as a robust, community-powered platform that lets you explore the frontier of generative AI from the comfort of your desktop. If you tried Pinokio in 2024 and found the setup tricky, the latest versions (3.6 and higher) have streamlined the experience. Pinokio's flurry of updates and platform improvements, along with the fast-growing library of one-click AI apps, has made running complex AI apps and models as easy as installing a phone app. Whether you want to generate videos, create art, clone voices, compose music, or deploy an AI agent, Pinokio can help you make it happen!

How to Get Started

1.     Download the latest Pinokio release at pinokio.computer

2.     Open the Discover page inside Pinokio to see featured apps. New arrivals often sit at the top.

3.     Install an app——like HunyuanVideoGP or WebUI by clicking the “Install” button. Pinokio does the rest.