---
title: "Rune Infer | Shirsendu Bairagi"
description: "High-performance, single-binary AI model gateway unifying local GGUF engines and cloud APIs with pre-flight VRAM budgeting."
canonical_url: "https://modesthumanbrands.com/work/rune-infer"
meta:
  "og:description": "High-performance, single-binary AI model gateway unifying local GGUF engines and cloud APIs with pre-flight VRAM budgeting."
  "og:image": "https://github.com/shba007/shba007/raw/main/logo.png"
  "og:image:height": "640"
  "og:image:width": "1280"
  "og:title": "Rune Infer | Shirsendu Bairagi"
  "twitter:description": "High-performance, single-binary AI model gateway unifying local GGUF engines and cloud APIs with pre-flight VRAM budgeting."
  "twitter:image": "https://github.com/shba007/shba007/raw/main/logo.png"
  "twitter:title": "Rune Infer | Shirsendu Bairagi"
---

> Index of every page on this site: [llms.txt](https://modesthumanbrands.com/llms.txt)

![Rune Infer preview](https://raw.githubusercontent.com/shba007/rune-infer/main/public/demo.gif)

[← All work](https://modesthumanbrands.com/#work)

# Rune Infer

High-performance, single-binary AI model gateway unifying local GGUF engines and cloud APIs with pre-flight VRAM budgeting.

- Rust
- GGUF
- WASM
- Axum

<dl>

<dt>Current version</dt>
<dd>v1.4.0</dd>

<dt>Last updated</dt>
<dd>Oct 15, 2026</dd>

<dt>Status</dt>
<dd>Open source</dd></dl>[Open Rune Infer →](https://github.com/shba007/rune-infer)

Overview

Rune Infer loads quantised models and serves them over a local socket or straight from the browser through WebAssembly. It features pre-flight VRAM budgeting, LRU model unloading, and process isolation so a failing model cannot crash the host.

## Capabilities

1. 01 ### VRAM Budgeting

   Pre-flight memory checks ensure models fit comfortably in available GPU/CPU memory.
2. 02 ### Zero-Crash Isolation

   Isolated execution engines guarantee that model exceptions never bring down the host gateway.
3. 03 ### Runs Anywhere

   Compiles to native binaries, container images, and WebAssembly modules.

## Releases

- v1.4.0Oct 15, 2026Current release
- v1.3.0Aug 20, 2026

[Next projectRune Kit →High-concurrency Rust host for the Model Context Protocol, running agent tools in memory-isolated wasm32-wasip1 sandboxes.](https://modesthumanbrands.com/work/rune-kit)

```json
{"@context":"https://schema.org","@graph":[{"@id":"https://modesthumanbrands.com/#website","@type":"WebSite","inLanguage":"en","name":"Shirsendu Bairagi","url":"https://shirsendu-bairagi.dev"},{"@id":"https://modesthumanbrands.com/work/rune-infer#webpage","@type":"WebPage","dateModified":"2026-11-10T13:31:56.779Z","datePublished":"2026-11-10T13:31:56.779Z","description":"High-performance, single-binary AI model gateway unifying local GGUF engines and cloud APIs with pre-flight VRAM budgeting.","name":"Rune Infer | Shirsendu Bairagi","url":"https://shirsendu-bairagi.dev/work/rune-infer","isPartOf":{"@id":"https://modesthumanbrands.com/#website"},"potentialAction":[{"@type":"ReadAction","target":["https://shirsendu-bairagi.dev/work/rune-infer"]}]}]}
```

## Sitemap

See the full [sitemap](https://modesthumanbrands.com/sitemap.md) for all pages.