Item detail

hwdsl2/self-hosted-ai-stack

Self-Hosted AI Stack packages Ollama, LiteLLM, AnythingLLM, Whisper, WhisperLive, Kokoro, embeddings, Docling, and MCP Gateway into a Docker Compose setup with local-first defaults, optional HTTPS, and multi-architecture support.

Score8.8
Popularity76.0
Riskconditional
TierGold
Score breakdown
Usefulness9.0
Novelty7.0
Momentum8.0
Maturity8.4
Open-source/build8.4
Evidence7.2
Workflow potential9.9
Setup ease6.4

Popularity is tracked separately. Support, ads, sponsorships, and tips never affect these signals.

Why it matters

Useful for people who want a private AI stack without spending a day wiring every container and environment variable by hand. It turns a messy integration project into a reproducible starting point for local or small-team deployments.

Who should use it

local AI usersself-hosting enthusiastssmall teams building private AI environmentsdevelopers who want an OpenAI-compatible local stack

Who should skip it

Skip if the source link, docs, or setup requirements do not match your workflow.

Risk explanation

The stack is private by default, but once you expose its services beyond localhost you need to verify generated API keys, TLS, and access controls so you do not accidentally publish internal model or document endpoints..

Evidence links

Closest alternatives / related signals

self-hostedollamalitellmmcp-gatewaylocal-ai