Quora’s Poe Launches Developer API for Multiple AI Models
▼ Summary
– Quora’s AI platform Poe released an API allowing developers to access various AI models for their applications without a separate fee, using Poe’s point-based subscription plans.
– API usage is tracked via points, with costs varying per model call (e.g., GPT-4o image generation costs 328 points for a 1024×1024 image).
– Poe offers tiered subscription plans ranging from $4.99/month (10,000 daily points) to $249.99/month (12.5 million monthly points), plus additional tokens at $30 per million.
– The API supports over 100 multimodal models (e.g., GPT Image 1, ElevenLabs) and works with tools like Cursor and Continue, compatible with OpenAI’s chat completion APIs.
– Poe plans to enable private bot API access and improve key management, while currently requiring manual model selection but may add budget features based on feedback.
Quora’s AI platform Poe has unveiled a new developer API that simplifies access to multiple artificial intelligence models, opening doors for innovative application integration. The service operates on a point-based subscription system rather than charging separate API fees, allowing developers to leverage various AI capabilities within their existing Poe plans.
Each API call consumes a predetermined number of points, with costs varying by model and task. For example, generating a standard 1024×1024 image using GPT-4o would deduct 328 points from a user’s balance. Poe offers tiered subscription options, starting at $4.99 per month for 10,000 daily points, scaling up to $249.99 per month for 12.5 million monthly points. Developers can also purchase additional tokens at a rate of $30 per million, with flexible top-up options available.
The API supports integration with tools like Cursor, Cline, Continue, and Roo, along with any platform compatible with OpenAI’s chat completion standards. Poe currently provides access to over 100 AI models, spanning text, voice, image, and video generation. Notable models include Imagen 4, GPT Image 1, Flux Kontext, Seedream 3.0, Veo 3, Runway Gen 4 Turbo, Kling 2.1, ElevenLabs, and Lyria, offering developers a broad toolkit for diverse applications.
Gareth Jones, Poe’s product lead for creators and developers, shared that the team is actively working on enhancements. “We’re enabling developers to deploy private bots built on Poe via the API and refining key management features for better usability,” he explained. Future updates may include budget management tools, depending on user feedback.
Beyond the API, Poe continues to expand its consumer-facing offerings. Earlier this year, the platform introduced streamlined tools for building AI-powered apps, including templates for server bots, prompt bots, and image generation bots. While developers currently manage model selection manually, Poe aims to refine the experience as adoption grows.
This move positions Poe as a versatile hub for AI development, catering to both individual creators and enterprise-level projects. With flexible pricing and a growing model library, the platform is poised to become a key player in the AI integration space.
(Source: TechCrunch)
