Fermyon
WebAssembly serverless platform with Serverless AI for LLM inferencing.
About
Fermyon builds developer tools for WebAssembly-based serverless computing, centered on Spin, its open-source framework for building Wasm microservices. With Spin 2.0, Fermyon introduced Serverless AI, letting developers run LLM inferencing directly from serverless apps with millisecond cold starts by time-sharing fractions of GPUs across applications.
The Spin SDKs support Python, Go, JavaScript, Rust, and .NET, and combine LLM access with a vector-enabled SQL database and key-value store for caching. Fermyon Cloud provides the hosted runtime, and the team are active contributors to WebAssembly standards through the W3C and Bytecode Alliance.
What it can do
- Serverless AI for LLM inferencing from Wasm apps
- Sub-second cold starts via GPU time-sharing
- Polyglot SDKs across five languages
- Vector-enabled SQL database and KV store
Setup
Install the Spin CLI, scaffold with spin new, build with spin build, and deploy to Fermyon Cloud with spin deploy.
Prerequisites
A Fermyon Cloud account for hosted deployment; local runs need only the CLI.
Use cases
Works with
Trust score — 8/10
Scored 8/10 (verified tier, metrics-v1). Evidence: Verified via vendor site/docs/GitHub and reputable coverage, wave 131; Trending score 78/100 from researcher assessment.
Sources
Every fact above was checked against these sources. Where a claim comes from the provider and couldn't be independently verified, the text says so.