feat(openrouter/openai/gpt-6-astra): add new models [bot]

[models-bot]

Auto-generated by model-addition-agent for openrouter/openai/gpt-6-astra.

---

Note

Low Risk
Adds a static model metadata file only; no application or routing logic changes.

Overview
Registers openai/gpt-6-astra on OpenRouter via a new provider YAML entry (auto-generated by model-addition-agent).

The definition sets chat mode with thinking enabled, a 1.05M context window, 128k max output, and text/image/PDF input. It documents prompt caching pricing, and capabilities such as function calling, tool choice, JSON/structured output, and prompt caching.

^{Reviewed by Cursor Bugbot for commit 15e0d99. Bugbot is set up for automated code reviews on this repo. Configure here.}

[cursor]

Cursor Bugbot has reviewed your changes using default effort and found 1 potential issue.

^{❌ Bugbot Autofix is OFF. To automatically fix reported issues with cloud agents, enable autofix in the Cursor dashboard.}

^{Reviewed by Cursor Bugbot for commit 63c6e8b. Configure here.}

[cursor]

Missing max_tokens override

Medium Severity

This fuller YAML advertises max_output_tokens of 128000, which exceeds the inherited OpenRouter max_tokens maxValue of 4096, and it does not override params.max_tokens. Callers would be capped well below the model's advertised output limit.

 

^{Triggered by learned rule: Override max_tokens maxValue to match limits}

^{Reviewed by Cursor Bugbot for commit 63c6e8b. Configure here.}

[github-actions]

/test-models