Poe Quora releases the API interface of a programmer with access to a bouquet of AI models

AI Quora platform Poe It was announced on Thursday that it releases the API interface, which allows programmers to easily access various models or bots for their own applications.

API does not require a separate fee. Instead, the use is tracked through existing subscription plans based on Poe, in which each connection of the model costs a fixed number of points. For example, producing a low quality image via GPT-4O in 1: 1 aspect ratio and size 1024 x 1024 would cost 328 points.

Today, POE plans include a plan of USD 4.99 per month (10,000 points per day), plan $ 19.99/MO (1 million points per month), plan 49.99 USD/MO (2.5 million points per month), plan 99.99 USD/MO (5 million points per month) and plan 249.99 USD/MO (12.5 million points per month).

Developers will also be able to buy additional points at a rate of 30 USD for 1 million tokens. Additional tokens are not in a fixed package, so customers can pay any dollar amount to get tokens based on this rate.

Through this API interface, programmers can, such as cursor, claline, string, roo and any other that work with API interfaces with an open chat. The platform currently provides access to over 100 models throughout the voice, text, image and video generation.

These multimodal models include Imagen 4, GPT Image 1, Kontext Flux, Seedream 3.0, Veo 3, Runway Gen 4 Turbo, Kling 2.1, Elevenlabs and Lyria.

“We are currently working on enabling programmers to download a private bot, which they have built on Poe and use it via the API interface. Plus, we are thinking about better management for programmers for the API product,” said Gareth Jones, POE product for creators and programmers, said TechCrunch over the phone.

While the API provides programmers by using different models, PoE also offers tools targeted at consumers. At the beginning of this year, the company introduced a new way to enable users to easily create AI powered applications. It also offers templates for creating server bots, fast bots and image generation bots.

At the moment, developers will have to manually choose and manage the use of models. Jones said that the company would consider adding budget management functions in the future, based on programmers' feedback.