Feedback and Feature Requests for Gamma API: Bugs, Documentation, and Enhancements for Consulting Reports

Hey Gamma team! 👋 I've been using the Gamma API quite heavily to generate consulting reports and strategy slide decks for the Korean market. Overall, it's a great product — especially the way Gamma detects layout structure from markdown patterns. Really impressive. I wanted to share a few bugs, feature requests, and questions based on production use. 🐛 Bugs / Documentation Issues 1. API parameter docs seem out of sync with the actual valid values • imageOptions.source: "stock" → returns 400 Only pexels, webFreeToUse, aiGenerated, etc. seem to work. • textOptions.amount: "concise" → returns 400 Only brief, medium, detailed, extensive appear valid. It would be very helpful if the docs listed the full enum values. 2. textMode: "preserve" doesn't fully preserve text In practice we see a few modifications happening: Fabricated numbers not present in the input (e.g. "82% utilization rate") Emoji being injected (e.g. "⚡ Complete in 4 weeks")

We currently work around this by adding strict rules at the top of additionalInstructions, but ideally preserve mode would strictly keep the original text unchanged. ⭐ Feature Requests 3. Support for ### (H3) markdown headings Right now H3 renders as plain text. For structured consulting reports we often need sub-heading hierarchy within slides. 4. Consistent heading font sizes ## headings sometimes render at slightly different sizes (~13–15pt) across slides within the same deck and theme. 5. Option to disable auto-generated section labels Gamma sometimes inserts large (~22pt) section labels on bullet-heavy slides that aren't present in the input. Something like: autoSectionLabels: false would be very helpful — especially when using preserve mode. 6. Higher slide limits / native batching support In practice the stable limit seems to be roughly:

Currently we generate multiple batches and merge them using python-pptx. Native support for 30+ slides or an API-level batch + merge workflow would simplify this a lot. 🙋 Questions 7. from-template API roadmap Currently supported parameters: gammaId, prompt, themeId, exportAs, imageOptions Are there plans to support additional parameters such as: textMode, additionalInstructions, textOptions? 8. Webhook callback support Right now we're polling every 12 seconds (up to ~50 times). Is there any plan to support completion webhooks? 9. Credit usage API Is there an endpoint to check: