Overview
Amallo is a tool from 41tunnels for remote access to a user's own Ollama installation. It keeps Ollama on the user's computer and provides an OpenAI-compatible endpoint that applications can reach without forwarding a router port or performing complicated setup. The site describes it as remote Ollama access that is OpenAI-compatible, with the user's models and machine reachable from anywhere.
Amallo runs in the menu bar or tray and finds Ollama on localhost:11434. It provides a base URL and an API key, with the OpenAI-compatible API enabled out of the box. Users can paste these credentials into any client that speaks the OpenAI API, such as Cursor, Aider, a script, or a compose file. The service is free, open source on GitHub, and licensed AGPL-3.0.
The site states that Amallo makes one outbound connection and keeps it open, so no static IP or open port is required and nothing on the router changes. Because the user's computer does the work, there is nothing to pay per message. Relay can be self-hosted; the source is on GitHub, builds to a single Go binary, and every release ships a container image, allowing Amallo to point at a user's own hostname.
Pricing
Free
Detailed plans have not been confirmed in our catalog. Check the official website for current limits and billing terms.
Visit WebsitePrices and limits may change. Confirm the currency, billing period, seat minimum and usage caps on the official website.
Use Cases
- Teams use Amallo to connect any OpenAI-compatible client to their own Ollama models.
- Teams use Amallo to point Cursor, Aider, scripts, or compose files at their local Ollama endpoint.
- Teams use Amallo to avoid router port forwarding when exposing Ollama.
- Teams use Amallo to keep Ollama on their own computer while accessing it remotely.
- Teams use Amallo to self-host Relay instead of trusting the public relay.
- Teams use Amallo to connect editors, agents, CLI tools, and custom scripts to Ollama.
- Teams use Amallo to run Ollama without per-message usage bills.
Features
AI assistant
Natural language interface
Workflow automation
Reviews
No reviews yet. Be the first to share your experience!
FAQ
Do I need a static IP or an open port?
No. Amallo makes one outbound connection and keeps it open, so nothing on your router changes.
What does it cost to run?
Your computer does the work, so there is nothing to pay per message.
Isn't this just ngrok or a Cloudflare Tunnel?
The site says it has the same shape as exposing a local port through a public relay, but it is scoped to Ollama rather than configured per tunnel. Amallo dials out on its own and provides a ready OpenAI base URL, so there is no tunnel URL to protect and no account to wire up.
Can I host Relay myself?
Yes. The site says this is the intended setup for anyone who would rather not trust the public relay. The source is on GitHub, builds to a single Go binary, and every release ships a container image; users can point Amallo at their own hostname.
Which clients work?
Anything that speaks the OpenAI API, unchanged, including editors, agents, CLI tools, and custom scripts.