design

Serge

A web interface for chatting with Alpaca through llama.cpp. Fully dockerized, with an easy to use API.
5,712 stars388 forksSvelteUpdated 11/21/2025100% free · open source
What it does

Serge gives you a ready‑made web UI and simple HTTP API for chatting with an Alpaca‑style LLM running on llama.cpp.

When to use it
  • You need a quick prototype UI for a custom LLM without building a front‑end from scratch.
  • Your team wants to expose a llama.cpp model via a REST endpoint for internal tools or bots.
  • You want to run an Alpaca‑style model locally (or on a private server) with minimal setup.
Ready-to-paste prompt
curl -X POST http://localhost:8008/api/chat \
     -H "Content-Type: application/json" \
     -d '{
           "messages": [
             {"role": "system", "content": "You are a friendly product manager."},
             {"role": "user", "content": "Draft a 3‑sentence pitch for a SaaS that auto‑generates meeting notes."}
           ]
         }'
Heads up: Serge only works with models compiled to the GGML format that llama.cpp supports; if you point `MODEL_PATH` to a regular PyTorch checkpoint the container will fail to start.
Saves to your device
Use with Claude
New

Skip the builder — one click puts this in Claude, Cursor, Antigravity and more.

✅ Light setup

Installs with a command or two; your AI agent can do it for you.

Try it instantly — no install
Claude Code
mkdir -p ~/.claude/skills/serge && curl -fsSL https://workflowstacks.com/api/skills/serge/claude-skill -o ~/.claude/skills/serge/SKILL.md
Open in another AI app

Opens the app with this repo with the prompt ready to go — no copy-paste needed.

Connect the whole catalog (MCP)
claude mcp add --transport http workflowstacks https://workflowstacks.com/api/mcp

Adds a WorkflowStacks connector to Claude Code: search and load any skill here by chatting.

How Serge works
Codeflow
Free to inspect

Serge is a medium Svelte project (~3.4k lines across 38 code files, plus 2 test files). Setup is light: runs with Docker. Last commit 10 months ago, Apache-2.0 license, has a test suite.

Size
Medium codebase
~3.4k lines · 38 code files · 28 min skim
Setup
Light setup
Runs with Docker.
Runs on
Docker
No API keys detected
Svelte 56%Python 30%CSS 5%Shell 3%TypeScript 3%
Where to start reading
  1. 1
    README.md
    Start here — what it does and how to install it
What's in each folder
web/Frontend / UI32 files
api/Backend / API25 files
scripts/Helper scripts3 files
.github/CI / automation (GitHub Actions)10 files
READMEHas testsThin docsCI checksDocker readyApache-2.0 licenseLast update 10 mo agoArchived — read only
Quick Actions
Details
Creator
serge-chat
Language
Svelte
Category
design
Published
3/19/2023

Are you the creator of this tool? Claim your listing → and earn 85% of every sale.