Fix OpenAI-compatible reasoning toggle for GLM 4.7/5 Turbo
fancive · ghdev · 2026-08-22
This PR resolves issue #42793 in anomalyco/opencode by updating ProviderTransform.reasoningToggle to recognize models with the npm package @ai-sdk/openai-compatible. When a reasoning toggle is declared in models.dev, the system now generates none/high variants that send thinking.type: "disabled" or "enabled" in the request body. This enables compatible endpoints like GLM 4.7 and GLM 5 Turbo to correctly toggle their thinking capabilities. The fix includes a regression test for this mapping and has been verified via end-to-end testing without affecting other provider behaviors.
More from coding & agent
- YC F26's Praxis launches AI agents to run law firm ops end to end — ycombinator · 2026-10-16
- Developer builds inter-agent messaging for his AI agents — complete with message undo — baditaflorin · 2026-10-16
- iOS 'dots' workflow: one Siri command turns screenshots into a background shopping and saving agent — nickbaumann_ · 2026-10-16
- Your widget tree is already a list of options: where Jev fits in Flutter — rseroter · 2026-10-16
- How OpenAI's Codex went from failed side project to 40M weekly users in under a year — danshipper · 2026-10-16
- Photon 2.0 launches unified messaging API for AI agents across iMessage and more — Scobleizer · 2026-10-16