Repository navigation
Replies: 1 comment
|
Are you interested in working together or using MetaCall or it's just marketing? |
0 replies
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
Hi MetaCall community,
I’m the author of TEN Framework:
https://github.com/TEN-framework/ten-framework
I wanted to briefly introduce the project here because I think TEN Framework and MetaCall share some similar architectural ideas, although they focus on different layers of the problem.
TEN Framework is an open-source framework for building real-time multimodal AI applications. It is mainly designed for use cases such as voice agents, video agents, STT, LLM, TTS, RTC, WebSocket transport, avatars, and other real-time AI pipelines.
The core idea of TEN is to let developers build applications by connecting independent extensions through a graph-based runtime. Each extension can be written in a different language, such as C++, Python, Go, or TypeScript/JavaScript. TEN Runtime then handles the lifecycle, message routing, threading model, and data flow between these extensions.
From my understanding, MetaCall focuses on direct function-level interoperability between languages. For example, language A can call a function implemented in language B through MetaCall’s runtime, loaders, ports, and type system.
TEN works at a slightly higher orchestration layer. Instead of exposing direct cross-language function calls, TEN lets extensions communicate through typed messages, commands, audio frames, video frames, and data streams. So the interaction model is more like a real-time graph of components.
A simple way to describe the relationship might be:
MetaCall:
language A calls a function from language B
TEN Framework:
extension A sends real-time messages or streams to extension B, and each extension may be implemented in a different language
So while the implementation details are different, the architectural concept feels somewhat similar: both projects try to make multi-language systems easier to build by introducing a common runtime layer between different languages or components.
Thanks!
All reactions