Demos: Deploying LLMs With WasmEdge
SEATTLE — Among the benefits of WebAssembly: it offers great promise as a runtime for large language models, due to its portability. WebAssembly (or Wasm, to its friends) allows developers to write code on any platform, in the language of their choice, without the need for recompilation.
“Wasm is — you can think of it as the Java bytecode,” said Miley Fu, devrel and founding member of Second State, in this episode of The New Stack Makers. “That’s because Java used to want to realize the dream of write once, run anywhere.”
In this On the Road episode of Makers, recorded at Open Source Summit North America in April, Fu demonstrated how to use WasmEdge, a runtime she created, with Gemma, an open source language model created by Google.
Watch the entire episode to see WasmEdge in action. And check out Fu’s two Open Source Summit presentations, both about WebAssembly and the Rust programming language: one on building serverless AI workflows and the other about creating AI agents.
