chalard.dev · private preview · one operator

Every model, one door.

A self-hosted LLM lab. Local models run on an old Xeon with no GPU; frontier models are proxied in. Chat, observe and build through a single OpenAI-compatible surface — behind Cloudflare Access.

curl https://api.chalard.dev/v1/chat/completions \
  -H "Authorization: Bearer $CHALARD_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"gemma4-e2b","messages":[{"role":"user","content":"สวัสดี"}]}'

Open Chat API Grafana Source

Lab = Models + Plumbing

The models are the soul. The plumbing keeps them reachable.

Three thin layers, each replaceable. Nothing here is clever; it is simply kept running.

01 · llama.cpp

Local inference

Gemma 4 E2B served from CPU RAM on a Proxmox VM. Bigger local models wait for a GPU; heavier work routes to hosted providers.

02 · LiteLLM

One API for all

Local and hosted models (GPT, Claude, Gemini, OpenRouter free tier) behind a single /v1. Keys, budgets and routing live in config.

03 · Cloudflare Access

Private by default

No open ports. Every hostname sits behind a tunnel and an identity check; the API takes a bearer key on top.

Design approach

Every request is visible. Every watt is counted.

Chat through Open WebUI

One interface for every model in the lab. Switch between a local Gemma and a hosted Claude mid-conversation; history, files and prompts stay on the box.

chat.chalard.dev →