Base
Hardened build
minimal ports, security patches applied at build time
Bifrost on Ubuntu 24.04
Azure
Artificial Intelligence (AI)
one OpenAI compatible endpoint in front of every model provider, packaged by cloudimg
Overview
Bifrost is a high performance open source LLM gateway. It sits between your applications and the model providers you use, exposing a single OpenAI compatible endpoint that fronts OpenAI, Anthropic, Google, Azure OpenAI, AWS Bedrock, Vertex, Groq, Mistral and Ollama, so existing application code usually needs only a base URL change. Behind that one endpoint it adds automatic failover between providers and keys, weighted load balancing, per key budgets and rate limits, semantic caching, MCP tool routing and a full request log with cost and latency for every call. Bifrost routes to providers using your own API keys and hosts no model itself, so it needs no GPU.
Why the cloudimg image
The cloudimg image pins the official upstream release binary by content hash and verifies it against the exact upstream commit before it ships. No model provider key is ever baked in: the image starts with zero providers configured and you add your own. There is no default password either, because an open LLM gateway is an open proxy for your inference spend. A unique dashboard password and a unique virtual key are generated on your VM at first boot, and the gateway refuses to start until first boot has proved an unauthenticated request is rejected on every route, including the admin API, the metrics endpoint and every inference path. Your applications authenticate with one credential using whichever header their SDK already sends. Paired with a verified deployment guide and 24/7 cloudimg support.
Common uses
- Give every application one governed, observable entry point to all your model providers
- Fail over between providers and keys automatically when one is rate limited or down
- Meter and cap spend per team or application with virtual keys, budgets and rate limits
See it running
Real screenshots taken while testing this image against its deployment guide.
More in Artificial Intelligence (AI)
ContextForge MCP Gateway - Hardened MCP Proxy
Artificial Intelligence (AI)
ContextForge MCP Gateway is IBM's open-source gateway, registry and proxy for the Model Context Protocol (MCP), the stan...
Crawl4AI
Artificial Intelligence (AI)
This image runs Crawl4AI, a fast, open source web crawler and scraper that converts web pages into clean, LLM-ready Mark...
Dify
Artificial Intelligence (AI)
This image runs Dify, a source-available platform for building and operating LLM applications.
Docling Serve Document Conversion API
Artificial Intelligence (AI)
Docling Serve is the HTTP API service around Docling, an open source toolkit that converts documents into clean, structu...
Evidently ML Observability Platform
Artificial Intelligence (AI)
This image runs Evidently, the open source platform for evaluating, testing and monitoring machine learning, large langu...
Gorse 0.5.11 on Ubuntu 24.04
Artificial Intelligence (AI)
The image installs Gorse as the official single all in one gorse-in-one binary (master, server and worker in one process...