Quora’s AI platform Po On Thursday, announced that it is issuing an API that allows developers to easily access various models or bots for their own applications.
API does not require a different fee. Instead, the existing point-based membership plans of the use are tracked through the use, where each model calls spend a fixed number of points. For example, the 1: 1 aspect ratio and 1024 x 1024 size will cost low quality image generation 328 digits through GPT-4o.
Today, POE plans include $ 4.99 per month scheme (10,000 marks per day), $ 19.99/Mo scheme (1 million marks per month), $ 49.99/mo scheme (2.5 million points per month), $ 99.99/mo scheme (5 million points per month), and $ 249.99/MO scheme (12.5 million marks per month).
Developers will also be able to buy additional marks at a rate of $ 30 for 1 million tokens. Ad-on tokens do not come in a certain package, so customers can pay any dollar amount to get tokens based on that rate.
Through this API, developers can power cursor, clin, continuing, roo and any other equipment that work with openi-compatible chat perfection API. The platform currently provides access to more than 100 models in voice, text, image and video generations.
These multimodal models include imagene 4, GPT image 1, flux contact, cidryam 3.0, VO3, runway gene 4 turbo, Kling 2.1, XIlabs and Leria.

Currently, we are working on allowing developers to take a private bot that they have created on POE and used that through API, we are thinking about better key management for developers for API products, “Po products for Garath Jones, Creators and Developers, explained a call for the products of Po, creators and developers.
Techcrunch event
San francisco
,
27-29 October, 2025
While API provides developers a way to use various models, POE also provides equipment for consumers aimed at consumers. Earlier this year, the company introduced a new way to allow users to easily create an AI-operated app. It also offers templates for the manufacture of server bots, prompt bots and image generation bots.
At the moment, developers will need to manually choose and manage model uses. Jones said that the company would consider adding budget management functionality in future based on developer response.

