Support for LLM APIs through the AI plugins #13480
-
|
Hello @dosu, Today, LLM providers offer a wide range of advanced capabilities through their APIs, including function/tool calling, agents, multimodal interactions, embeddings, reasoning models, advanced streaming, context management, and more. Are there any plans in the APISIX roadmap to introduce additional AI plugins or enhance the existing ones to provide native support for more LLM providers and the different APIs and features they offer? Given the rapid pace of innovation in the AI space, how does the APISIX team plan to keep up with new providers and API capabilities without having to build and maintain provider-specific plugins for each one? Best regards. |
Beta Was this translation helpful? Give feedback.
Replies: 2 comments 7 replies
-
|
@dosu |
Beta Was this translation helpful? Give feedback.
-
|
thanks |
Beta Was this translation helpful? Give feedback.
APISIX doesn't maintain a detailed, version-specific public roadmap with release timelines. Instead, it has a lightweight Vision-and-Milestones.md file in the repository root that outlines general directions for "the next 6 months" [1], including items like Gateway API support, service mesh, documentation improvements, and more plugins for cloud services.
The project follows a community-driven development model [2] — features tend to emerge organically from community needs and contributions rather than from a rigid, top-down release schedule. There are no public GitHub project boards or milestone trackers tied to specific upcoming versions.
In practice, the best way to track what's comin…