Fetching from the wire…
01
02
03
04
05
06
07
08
Public story · 2026-07-16 · high
A Claude agent provider joins the plugin in public preview, available to Pro tier and up, per GitHub's July 14 changelog.
Why now: The BYOK expansion landed in GitHub's July 14 changelog, covered in the July 16 briefing.
GitHub updated Copilot for JetBrains on July 14 to accept any OpenAI-compatible endpoint with your own API key, per its changelog. That's a shift in what you pay GitHub for. The plugin becomes a routing layer over models you pick, not a closed pipe to GitHub-provisioned inference.
A Claude agent provider joined the same update in public preview, available to Pro tier and up, according to the changelog. Copilot for JetBrains spans three paths as of this release: GitHub's own models, a self-hosted or third-party OpenAI-compatible backend, or Claude through the preview provider.
The changelog entry doesn't list minimum plugin or IDE versions, and it doesn't say whether there are model compatibility limits on the custom endpoints. If you're planning a team rollout, test that first, especially for anything tool-calling or multi-step. An OpenAI-compatible API surface doesn't guarantee every self-hosted model behaves the same way inside Copilot's agent flows.
The Claude provider is preview-only and gated to Pro tier and up, per the changelog's tier callout. Whether free and Business seats get BYOK without the Claude path isn't stated, so confirm your tier before rolling this out.
Each link below shares sources, entities, or timing with this story.
July 9, across VS Code, Visual Studio, Copilot CLI, the cloud agent, github.com, GitHub Mobile, JetBrains, Xcode, and Eclipse. Sol is the high-reasoning tier at $5/1M in, $30/1M out, gated to Pro+/Max/Business/Enterprise. Terra is the balanced default at $2.50/$15. Luna is fas...
GitHub shipped it July 28 across Pro through Enterprise, reachable from VS Code, Visual Studio, Copilot CLI, the cloud agent, JetBrains, Xcode, and Eclipse, with text and image inputs and low/medium/high reasoning effort, billed at provider list pricing rather than a fixed mul...
43.3% on Frontier-Bench v0.1. Opus 4.8 scored 18.7%. That's not an incremental bump, that's the same benchmark with a different shape of answer. Anthropic released Claude Opus 5 on July 24 at $5/$25 per million input/output tokens, exactly half of Fable 5's $10/$50, while matc...
JetBrains released Junie Local on August 24 (JetBrains blog). You type /local inside Junie, it pulls about 20GB of 4-bit weights, starts a local server, and from that point there are no tokens, no quota, and no code leaving the machine. Free. The hardware bar is real and steep...
An open-weight Chinese frontier model is now a dropdown option in Microsoft's coding product. That happened before anyone finished characterizing what the model does. GitHub's changelog dated August 6 makes Kimi K3 generally available across Copilot Pro, Pro+, Max, Business an...
The open-source coding agent space just got its first credible frontrunner. OpenCode, built by Anomaly, launched this week and immediately became the top technical story on Hacker News with 802 points and 359 comments — the kind of signal velocity that separates real developer...
MindPattern daily
One email a day at 7 AM. Sources and a take on every story. Unsubscribe anytime.