Open Source & Self-Hosted

Own your AI
infrastructure

A self-hosted, open-source platform for deploying, managing, and scaling conversational AI across your organization. Multi-model. Enterprise-grade. Yours to control.

chat.yourdomain.com
Claude Sonnet 4
Agents
Chat

Built for teams that take AI seriously

Self-hosted & open source

Deploy on your infrastructure. Full data sovereignty — no vendor lock-in. MIT licensed and auditable.

Enterprise team management

Per-user and per-team budgets, identity federation with Azure AD and Keycloak, granular model access control.

Multi-model unified gateway

One interface for Anthropic, OpenAI, Google, and local models. Switch providers mid-conversation. No rewiring.


Up and running in minutes

Three steps from zero to a production AI platform.

01

Deploy

Clone the repo and run Docker Compose. The full platform spins up with PostgreSQL, Redis, Meilisearch, and authentication — all pre-configured.

bash ~/SecondStack
# Clone & initialize
git clone https://github.com/
  SecondStack-AI/SecondStack
cd SecondStack
./init_platform.sh
02

Connect providers

Add your LLM provider API keys through ControlTower. Supports Anthropic, OpenAI, Google Gemini, and local models via vLLM. Configure models, pricing, and capabilities in one place.

A
anthropic
O
openai
G
google gemini
L
local (vLLM)
03

Invite your team

Set up teams, assign budgets, and connect your identity provider. Users get instant access to a multi-model chat interface with full cost tracking.

Engineering $342 / $500
Product $89 / $200
Support $21 / $100

Everything your team needs

A complete conversational AI platform, not just a chat interface.

Chat App

Multi-model chat

Switch between Claude, GPT, Gemini, and local models mid-conversation. Grouped by capability — chat, reasoning, image generation. Your team picks the right model for each task.

Chat Models
Reasoning Models
Image Generation
ControlTower

Admin dashboard

Centralized model management, provider configuration, system prompts, and deployment controls. One panel for your entire AI stack.

admin.yourdomain.com/dashboard/admin?section=models
Control Tower
A

Models

Model Display Name Access Status Health Order Actions
A
claude-sonnet-4-6
Claude Sonnet 4.6 All Users active 2/2 OK
A
claude-opus-4-6
Claude Opus 4.6 All Users active 3/4 OK
O
gpt-4.1
GPT-4.1 Engineering active 2/2 OK
G
gemini-2.5-flash
Gemini 2.5 Flash All Users active 1/1 OK
A
agent-claude(agent)
Agent Claude 2 Teams active N/A
O
gpt-image-1(image_generation)
GPT Image 1 Hidden disabled

Users

User Email Role Teams Last Active Actions
JC
Jane Chen
jane.chen@acme.com Admin Engineering, Platform 2 min ago
MR
Mike Rodriguez
mike.r@acme.com User Engineering 14 min ago
SP
Sarah Park
sarah.p@acme.com User Design, Marketing 1 hr ago
AK
Alex Kim
alex.kim@acme.com Manager Engineering, QA 3 hrs ago
LW
Lisa Wang
lisa.w@acme.com User Marketing 1 day ago

Providers

A
Anthropic Direct API Key
connected
Models4
Latency142ms
Uptime99.9%
O
OpenAI API Key
connected
Models3
Latency98ms
Uptime99.8%
AWS
AWS Bedrock IAM Role
connected
Models2
Latency167ms
Uptime99.7%
G
Google AI Service Account
degraded
Models1
Latency312ms
Uptime97.2%

Teams

E
Engineering 12 members
active
Budget$2,400/mo
Used$1,847
Models6
D
Design 5 members
active
Budget$800/mo
Used$612
Models4
M
Marketing 8 members
active
Budget$600/mo
Used$423
Models3

Groups

GroupSourceMembersTeamsActions
engineering-allAzure AD24Engineering, Platform
design-teamAzure AD8Design
marketing-allManual12Marketing

Analytics & Logs

Total Requests48,291+12.4%
Total Spend$3,847+8.2%
Avg Latency156ms-5.1%
Error Rate0.3%-0.1%
TimeModelUserTokensLatencyStatus
14:32:01claude-sonnet-4-6jane.chen2,8471.2s200
14:31:58gpt-4.1mike.r1,2040.8s200
14:31:45claude-opus-4-6alex.kim5,1023.4s200
14:31:32gemini-2.5-flashsarah.p8920.4s200

General Config

Platform Name
Acme AI Platform
Default Model
claude-sonnet-4-6
Max Tokens
32,768
Auth Provider
Authentik (OIDC)
Sync Schedule
Every 5 minutes

Notifications

Budget Alerts
enabled
Health Alerts
enabled
Slack Webhook
https://hooks.slack.com/...4xK
Email Recipients
admin@acme.com, ops@acme.com

MCP Servers

ServerTypeToolsStatusActions
filesystemstdio6 toolsactive
web-searchsse2 toolsactive
databasestdio4 toolsdegraded

Guardrails

RuleScopeActionStatus
PII DetectionAll ModelsBlock & Alertactive
Prompt InjectionAll ModelsBlockactive
Code ExecutionAgent ClaudeWarnactive

Localized Content

LocaleStringsCoverageStatus
English (en)247100%default
German (de)24197.6%active
Japanese (ja)18976.5%partial

Config & Deploy

LiteLLM Config
syncedLast deployed 4 min ago
Chat App Config
pending2 changes waiting
Guardrails Config
syncedLast deployed 1 hr ago
vLLM Config
syncedLast deployed 2 hrs ago
Cost Management

Budget & spend controls

Per-user, per-team, and per-API-key budgets. Threshold alerts, spend forecasting, and usage analytics down to the request level.

Budget Overview February 2026
Total Platform Spend $1,247 / $2,500
Remaining: $1,253
Engineering $642 / $1,000
Chat App $380
Agent Claude $198
API Keys $64
Product $289 / $500
Chat App $245
Agent Claude $32
API Keys $12
Marketing $198 / $200
98% of budget used
Chat App $186
Agent Claude $12
API Keys $0
Support $118 / $800
Chat App $89
Agent Claude $22
API Keys $7
Marketing team is approaching budget limit (98%)
View Details →
Agent Claude

Agentic code execution

Agent Claude runs in sandboxed containers with 50+ skills — file operations, browser automation, image generation, and persistent memory across sessions.

Security

Enterprise identity

Built-in OAuth with Authentik. Federation with Azure AD and Keycloak. Automatic group sync and role-based access control.

Infrastructure

Search & retrieval

Meilisearch-powered full-text search across threads and messages. Find any conversation instantly. Document extraction with Docling and Kreuzberg.


Open source. Deploy anywhere.

SecondStack is MIT licensed. Run it on your hardware or let us help.

Community

Free

Open source, forever

Full platform for teams getting started with self-hosted AI.

  • All platform features
  • Docker Compose deployment
  • Community support via GitHub
  • Multi-model chat interface
  • Team & budget management
  • Identity federation
Deploy Now

Enterprise

Custom

Tailored to your needs

For organizations requiring dedicated support and advanced deployment.

  • Everything in Community
  • Kubernetes deployment support
  • Priority support & SLA
  • Custom integrations
  • Deployment assistance
  • Security review & hardening
Contact Us

Frequently asked questions

Can't find your answer? Open an issue on GitHub.

Which LLM providers does SecondStack support? +
SecondStack supports all major providers through LiteLLM: Anthropic (Claude), OpenAI (GPT), Google (Gemini), and 100+ others. You can also run local models via vLLM integration.
How does self-hosting work? +
Clone the repository, configure your environment variables, and run the initialization script. Docker Compose orchestrates all services — PostgreSQL, Redis, Meilisearch, authentication, and the application layer. The entire platform runs on a single machine or cluster.
What are the system requirements? +
A Linux server with Docker and Docker Compose. Minimum 4 CPU cores and 16GB RAM for a basic deployment. Storage scales with usage — primarily chat history and document attachments.
Is SecondStack really free? +
Yes. SecondStack is MIT licensed — free to use, modify, and distribute. You pay only for your LLM provider API costs and your own infrastructure. No hidden fees or usage-based pricing from us.
How does team and budget management work? +
Admins create teams and assign per-user or per-team budgets through ControlTower. Budgets can be set for Chat App usage, Agent Claude, and API keys separately. Threshold alerts notify when spending approaches limits.
Can I use local models instead of cloud APIs? +
Yes. SecondStack includes vLLM integration for running open-source models locally. You can mix local and cloud models — some teams use local models for sensitive data and cloud APIs for general use.

Ready to own your AI infrastructure?

Deploy SecondStack in minutes. Open source, self-hosted, and built for teams that demand control.