Introducing "Poe" and it's Applications

[smfadmin (OP)]

Hi everyone,

Over the last two years, we have built Poe as a unified chat interface to all AI. We now provide millions of users with access to 100 of the best text, image, video, and audio models from leading AI companies, along with millions of bots built on top of them, all in one place.

We have learned from this experience that while chat is the right interface for many people and for many use cases, there are also many cases where a visual UI is a better interface, particularly when a single workflow involves the use of multiple models made by different companies — access that Poe can uniquely provide.

Today we are introducing Poe Apps, which make it easy to build visual interfaces on top of any combination of the existing models on Poe and custom logic expressed in JavaScript. Poe Apps can run side-by-side with chat, or be entirely visual. Thanks to the integration with Poe, when these apps need to make potentially-expensive model calls, they will use the user's existing points, so that as a creator, you don't have to risk ending up with huge API bills, or having downtime when your app hits limits.

You can create a Poe App without writing any code, thanks to App Creator, which we built on top of Claude 3.7 Sonnet. Or, if vibe coding is not for you, you can take full control of the underlying HTML and JavaScript. Either way, we'll be giving distribution to the best apps within Poe through the existing systems we have built out for bots. This means our millions of users are all your potential audience, with no extra work required on your part.

Here are some examples we created to better illustrate this product:

Chibify: An app that transforms one or more photos into 3D anime style art, using GPT-4o and FLUX-Pro-1.1.

MagicErase: An app that removes unwanted objects from images, using Bria-Eraser.

MemeGenBattle: An app that lets you compare memes generated by LLMs of your choosing.

These are just a few examples, but we expect a wide diversity of products to be created because Poe Apps can build on all the models on Poe.

This includes the newest reasoning models, including o3-mini, Claude 3.7 Sonnet, and DeepSeek R1, multimodal text generation models like Gemini 2 and GPT-4o, image generation models like FLUX Pro 1.1, Imagen 3, Ideogram v2, Playground v3, and DALL-E 3, video generation models like Veo 2, Pika 2.0, Hunyuan Video, Kling Pro 1.5, and Hailuo Live, and audio generation models like ElevenLabs and Cartesia.

Starting today, Poe Apps are available on web, with iOS and Android support coming in the near future. You can get started with App Creator here.

This is an early launch and we have a long roadmap ahead to give creators even more power, including the ability to earn money directly from their apps. We are excited to see what you all create, and we are excited to see how much better we are able to make Poe Apps as the models continue to get better at writing code this year.

Please share any feedback with us at feedback@poe.com.

Thank you!
The Poe Team

This email was sent by Poe (605 Castro Street, Mountain View, CA 94041).