SecondStack launched SecondStack v2.5.0, which includes support for the new GPT-6 Astra model from OpenAI.

SecondStack is a self-hosted enterprise AI platform. Your teams get chat, coding, and agents on the models you choose. Your platform team gets one gateway with central keys, budgets, routing, and audit. All of it runs on infrastructure you own. Conversations, attached documents, and usage data never leave your perimeter. There is no per-seat pricing, so cost does not scale with headcount. What's inside: • Chat: a multi-model assistant for every employee, with knowledge collections and custom prompts • Code: AI coding assistance inside your own environment • Agent: automation across your internal systems and data • Gateway: one enterprise LLM API gateway across providers, with per-user, per-team, and per-key budget enforcement • Admin: central controls, guardrails, and usage visibility Built for platform, security, and engineering leaders who need AI adoption without handing data or budget control to a SaaS vendor. Deploy on-premises or in your own cloud. Own Your AI Infrastructure.