mediapipe-rs VS SSVM

https://wasmedge.org/docs/contribute/plugin/intro

Some examples:

WasmEdge’s Tensorflow Lite plugin enables lightweight (1/10 of Linux containers) and fast (native GPU) AI inference apps for Tensorflow Lite models. https://wasmedge.org/docs/develop/rust/ai_inference/tensorfl...

WasmEdge’s PyTorch plugin enables lightweight (1/10 of Linux containers) and fast (native GPU) AI inference apps for PyTorch models. https://wasmedge.org/docs/develop/rust/ai_inference/pytorch/

Through the new WasmEdge plugin system, the community are adding support for libraries like zlib, OpenCV, tesseract and ffmpeg etc. Putting these together, we are supporting complex AI libraries such as Mediapipe on WasmEdge! https://github.com/yanghaku/mediapipe-rs

The Mediapipe story is esp interesting since WasmEdge is now adapted as a runtime for stream data process in products like Fluvio, Redpanda, YoMo, RisingWave and others. Mediapipe support allows developers to add AI into the stream data pipeline. https://github.com/xxchan/fluvio/pull/1

WasmEdge’s eBPF plugin enables developers to create secure and containerized eBPF tools and controllers in Kubernetes environments.

WasmEdge GitHub address: https://github.com/WasmEdge/WasmEdge.

I see that they recently changed the intro of WasmEdge on Github [1] to " WasmEdge is the easiest and fastest way to run LLMs on your own devices. "

Since it's a wasm runtime capable of many things I find bizarre that they now start describing it with a ultra-specific use case

- [1] https://github.com/WasmEdge/WasmEdge

Mmm…

The wasm-nn that this relies on (https://github.com/WebAssembly/wasi-nn) is a proposal that relies of arbitrary plugin backends sending arbitrarily chunks to some vendor implementation. The api is literally like set input, compute, set output.

…and that is totally non portable.

The reason this works, is because it’s relying on the abstraction already implemented in llama.cpp that allows it to take a gguf model and map it to multiple hardware targets,which you can see has been lifted here: https://github.com/WasmEdge/WasmEdge/tree/master/plugins/was...

So..

> Developers can refer to this project to write their machine learning application in a high-level language using the bindings, compile it to WebAssembly, and run it with a WebAssembly runtime that supports the wasi-nn proposal, such as WasmEdge.

Is total rubbish; no, you can’t.

This isn’t portable.

It’s not sandboxed.

If you have a wasm binary you might be able to run it if the version of the runtime you’re using happens to implement the specific ggml backend you need, which it probably doesn’t… because there’s literally no requirement for it to do so.

There’s a lot of “so portable” talk in this article which really seems misplaced.

WasmEdge isn't there yet: https://github.com/WasmEdge/WasmEdge/issues/1122#issuecommen...

Example 1: Analyze the content and risks of each commit in the PR. Then make a summary. https://github.com/WasmEdge/WasmEdge/pull/2394#issuecomment-...

There has also been a few "cloud-native" runtimes based on WASM, like WasmEdge but there's a few others (can't remember their names!)...