Build AI Companions that understand and complete tasks for your users in apps. Our AI Agent foundation models outperform GPT-4o in function calling and support use cases for shopping, travel booking, video streaming, video conferencing apps, and much more!
Congratulations on the launch, it looks very impressive. I have already registered on your website and reviewed your developer documentation, but I still have some confusion about how to use it specifically.
It seems like the playground is in JSON mode, but in the developer documentation, it appears to be a way to route the API through LLM. So, is the usage of this that I need to register all my APIs and their corresponding descriptions in one place, and then integrate your services? After posting the user's natural language input to your service, will your service determine which API needs to be called as a function call?
I still haven't figured out how to practically use your product, and I would like to know if there are any developer-friendly examples available.
@brandon_cccc
Hi Brandon, I am an Engineer at Nexa AI. Thank you for your interest and congratulations on registering and reviewing our developer documentation!
To address your questions:
Playground and API Integration: Our playground and the open API interface are actually the same. The playground serves as a demo to show how our model determines which function to use based on different natural language inputs.
Model Training for Use Cases: Currently, we have trained four different models for four specific use cases. These models will return pre-determined function names and parameters based on the user's input. As a developer, you are responsible for implementing these functions.
Custom APIs and Model Training: If you have custom APIs and want to train a model specifically for your use case, you can click "contact us" on the Pricing table. We have the capability to train a model tailored to your specific needs.
For example, in the shopping category, as a developer, you might implement functions like see_recommended_products and search_products. Our model will return the function name and parameters based on the user's query, allowing you to directly call the functions you have written.
I hope this clarifies how to practically use our product. If you have any further questions or need additional examples, please don't hesitate to reach out!
@brandon_cccc For faster support and detailed guidance, we welcome you to join our Discord grouphttps://discord.com/invite/thRu2.... We’ll be more than happy to answer any further questions and provide assistance there.
Hi Alex,
Congratulations on launching Octoverse! 🚀It’s clear that Octoverse can significantly enhance user experience and drive business growth.
Can't wait to see how it helps transform the way we interact with technology.
Best,
Sanja
Octoverse
Flex-Worthy Templates
Octoverse
NexaSDK for Mobile
AI HomeDesign
NexaSDK for Mobile
NexaSDK for Mobile
Riffo
Octoverse
Octoverse
Octoverse
Founder Reports