Self-hosted alternatives to Compute Cheap
200 self-hosted apps like Compute Cheap, closest match first.
Top picks
- B

Beam
LaunchBeam is a cloud platform for running GPU workloads, sandboxes and task queues for AI agents and applications. You can run custom model inference on serverless GPUs, start persistent Linux virtual machines, and create sandboxes that boot qui
Open sourceSelf-hosted
- K
Kalavai Client
LaunchKalavai is an open-source platform that pools compute from spare capacity, such as idle GPUs on your own machines, so you can run larger AI workloads. It acts as a control plane for computing resources wherever they are, including local mac
FreeOpen sourceSelf-hosted
- I
Income Generator
LaunchA set & forget, super-lightweight passive income manager. Deploy once, scale with multi-proxy support. Earn anywhere. Orchestrate via CLI, TUI or WebUI.
FreeOpen sourceSelf-hosted
- N
NadirClaw
LaunchSource-available LLM router and AI cost optimizer. Routes simple prompts to cheap or local models and complex ones to premium models, automatically. Drop-in OpenAI-compatible proxy for Claude Code, Codex, Cursor and OpenClaw. Self-hosted, n
FreeOpen sourceSelf-hosted
- H

Helix
Launch♾️ Private Agent Fleet with Spec Coding. Each agent gets their own GPU-accelerated desktop. Run Claude, Codex, Gemini and open models on a full private AI Stack ♾️
FreeOpen sourceSelf-hosted
- G

Guaardvark
LaunchSelf-hosted AI studio on your own GPU: chat with your files (RAG), screen and browser agents, coding swarms, LoRA training, an MCP server for Claude Code and Cursor, and local video, image, music and voice generation. No telemetry.
FreeOpen sourceSelf-hosted
More like Compute Cheap
OmniKey-AI-Unified-Key-Manager
A self-hosted LLM proxy and failover gateway supporting OpenAI & Gemini…
Self Host N8n On GCR
Self-host n8n on Google Cloud without the subscription fees or server headaches
About Compute Cheap
Compute Cheap
The world’s cheapest GPU compute
Compute Cheap rents prepaid, interruptible GPU capacity on NVIDIA H100, H200 and B300 hardware in the US and EU, aimed at teams running model training, inference and batch jobs that can tolerate interruptions. You request a block of capacity, the team confirms availability for that exact request, and then you run your workload. Because capacity is interruptible, it can be reclaimed when demand spikes, so it suits fault-tolerant work with checkpointing. Requests have a 2,000 GPU-hour minimum, roughly eight GPUs for ten days, and once availability is confirmed you prepay half of the block, with the rest billed as usage completes. The site publishes a leaderboard comparing its rates with other GPU clouds, a pricing API with live prices as JSON, a status page and case studies. Reserved capacity is not currently offered for new requests.
- Price
- Paid
- Category
- Finance
- Website
- compute.cheap



