proomt

Search

Search posts, papers, and topics

All posts

InfoQCraig Risi2 min readrelease notesintermediate

GitLab Duo Expands Self-Hosted AI Options Through Microsoft Foundry

Read on infoq.com

Summary

GitLab Duo Self‑Hosted now integrates with Microsoft Foundry, allowing AI requests to flow through a self‑managed GitLab AI Gateway to Azure‑hosted OpenAI, Anthropic, Meta, or Mistral models. This gives enterprises control over model choice, data residency, and credentials, while adding operational responsibility for model lifecycle and compatibility.

  • GitLab Duo Self‑Hosted can use a self‑managed AI Gateway to route requests to models deployed on Microsoft Foundry (OpenAI, Anthropic, Meta, Mistral).
  • Per‑feature model selection lets teams pick different models for code suggestions, agentic workloads, or high‑volume tasks while keeping traffic inside their Azure environment.
  • Engineering teams must now manage model deployment, capacity, networking, credentials, and compatibility between Foundry catalog and GitLab’s supported matrix.
  • The integration decouples GitLab’s DevSecOps layer from any single foundation model, making it a model‑agnostic control plane.

Enterprises needing data sovereignty or custom AI model control should care because GitLab now lets them host models on Azure while preserving their DevSecOps workflow.

5/10
devopssecurityarchitectureml infra

Related reading

  1. Best GitLab Alternatives for Enterprise Teams (2026)

    The article is a vendor‑authored marketing piece that lists criteria for picking a GitLab alternative and then promotes CloudBees Unify as a cross‑tool governance layer, with brief mentions of other tools (GitHub, Jenkins, Azure DevOps, Harness, CircleCI). It contains a few quoted survey percentages but no technical deep‑dive, code, architecture diagrams, or independent evaluation.

    Codeshipcloudbees.com10 min

  2. GPT-6 Sol and Luna now available on AI Gateway

    Vercel adds OpenAI’s GPT‑6 Sol and Luna models to its AI Gateway. Sol targets complex, high‑quality coding and professional workflows; Luna is a cheaper, high‑throughput option. Both are accessible via Vercel’s AI SDK, OpenAI‑compatible APIs, and coding agents, with example TypeScript usage and CLI setup instructions.

    Vercelvercel.com1 minrelease

  3. GLM 5.3 FlashX now available on AI Gateway

    Vercel AI Gateway now offers the GLM‑5.3‑FlashX model, a fast (~200 tps) multimodal coding LLM. The post includes a TypeScript streaming example, notes the model’s fit for coding agents and interactive tools, and lists AI Gateway features (unified API, usage tracking, retries/failover, custom reporting, key budgets, routing rules) with no platform fee.

    Vercelvercel.com1 minrelease

  4. Aperture GA: Building a home(lab) for agentic AI

    Aperture, Tailscale’s AI gateway, is now generally available. It ships with starter tokens, in‑app token purchases, new Model Context Protocol (MCP) endpoints for adding Tailnet nodes and invoking Tailscale SSH, and an upgraded chat UI that supports Projects (shared context, tool access, node groups) and configurable default tool permissions. All agent actions respect Tailscale’s unidirectional A…

    Tailscaletailscale.com4 minreleaseHN41

  6. GPT-Live 1 now available on AI Gateway

    Vercel AI Gateway now offers OpenAI’s GPT‑Live 1, a full‑duplex voice model that can listen and speak simultaneously. The SDK lets you start a voice session and optionally delegate work to any text model, with code examples for both patterns.

    Vercelvercel.com1 minrelease