Building Serverless AI Workflows with Wasm+Rust
summary
Claims a portable serverless AI stack using WasmEdge + Rust that runs locally with zero Python dependency, compiles Rust functions to Wasm and ships a single wasm binary that runs on Mac/Intel/ARM and various GPUs. Demo shows a one-line install, a local API on localhost:880 and a working TinyLlama inference with binaries under <30MB binary. Platform compiles Discord bot functions, integrates Hugging Face models and is OpenAI API compatible while retaining sandboxed security.
AI-written summary · about
chapters (12)
- 0:00Serverless AI with Wasm+Rust
- 1:05AI workflow challenges
- 10:02Wasm+Rust for AI performance
- 14:05WebAssembly for servers
- 18:46Wasm for AI inference
- 25:07Dev vs Ops with Wasm
- 28:53WASI-NN and GPU for AI
- 30:22Demo: Discord bot
- 36:01Demo: TinyLlama locally
- 37:36Serverless AI use cases
- 40:31AI frameworks beyond LLMs
- 41:15Open source contributions
related talks
41mCreating Versatile AI Agents Through Wasm+Rust
55mFast And Portable LLM Inference With WebAssembly And Rust
25mLeveraging Wasm for Portable AI Inference Across GPUs, CPUs, OS & Cloud-Nativ...
41mWebAssembly Based AI as a Service
40mWebAssembly for AI Infra : A Lightweight, Fast, and Secure Alternative Approach
01h 04mDevelop, Build & Deploy Lightweight AI Services with WebAssembly
27mWasm Is Becoming the Runtime for LLMs
10mLightning Talk: Scaling AI With Wasm and Edge Computing