CICI/CD News
Platform engineering

GitHub closes its Models playground to new customers and routes new work to Azure AI Foundry

June 19, 2026githubgithub-modelsazure-ai-foundrydeprecationaideveloper-workflow
Maya Okonkwo · Site Reliability EngineerGitHub closes its Models playground to new customers and routes new work to Azure AI Foundry

GitHub is retiring GitHub Models, the free in-platform AI playground it introduced in 2024, and any pipeline or experiment wired through its API now has a migration path attached to it. Per DevOps.com, the feature is closed to new customers as of June 16: organisations that have not used it before cannot start. Existing customers with active usage "can continue using the playground, API, and models as usual," and GitHub is directing developers who need model access for new projects to Azure AI Foundry.

What is closed and what is not

The cutoff is on enrolment, not on running workloads. New customers are out. Existing customers keep playground, API and model access — the wording DevOps.com pulls is "as usual," which is the part to read carefully. As-usual coverage carries no end date in what is reported, and no parallel commitment that the underlying models, rate limits or routing stay frozen. A grandfathered surface that works today is not the same as a surface a platform team can safely build a new pipeline against.

GitHub Models gave a single API key access to Meta's Llama 3.1, OpenAI's GPT-4o and GPT-4o mini, Cohere's Command, and Mistral Large 2 — a deliberately broad menu through one credential. That convenience is what disappears for new starters. Azure AI Foundry, the recommended next step per DevOps.com, is a different surface: separate account path, separate billing and a different operational model from "the key already lives in my GitHub org."

The operational read

For teams already pulling models through that single key in CI — release-note drafting, PR summarisation, test triage — the immediate task is an inventory of where the credential is referenced and what would break if "as usual" stops being literally true. There is no published timeline tying that surface to an end date. There is also no commitment that the model menu behind it stays the same. For teams that were planning to start, the door is already shut and the work to evaluate Azure AI Foundry has effectively moved up the backlog.

The retirement is signposted. The deadline is not.

Source: DevOps.com (devops.com)

Related
Code quality & testing

GitHub Code Quality gets a GA date — and a seat inside the merge gate

GitHub has set July 20, 2026 as the general-availability date for Code Quality, a native engine that wires maintainability and reliability checks, code-coverage tracking and merge-blocking quality gates into the pull-request flow. For pipelines, the change shifts a class of checks from third-party SaaS into the same place where the merge decision lives.

June 17, 2026Security & supply chain

Docker Content Trust gets a sunset date. The harder question is what you sign with next.

Docker has published a formal retirement plan for Docker Content Trust and the Notary v1 service at notary.docker.io, ten years after DCT shipped. The migration is mostly mechanical — the strategic question, about whether anyone downstream was actually verifying anything, is the part the guide cannot answer for you.

June 16, 2026Platform engineering

Block runs its coding-agent fleet from Slack. The chat is the easy part.

The New Stack describes how Block manages and supervises a fleet of AI coding agents from Slack, framing the real problem as operating across many services rather than inside a single repository. For a CI/CD practitioner, the interesting shift is from prompt quality to the control plane around the agents.

June 19, 2026

Turn this into your pipeline. Build it on Buddy.

Start free