Gryphe/Opus-4.6-Reasoning-24k

Opus-4.6-Reasoning-24k While playing with reasoning-based finetunes I ended up building a small pipeline to aggregate, verify, normalize and deduplicate all the Claude Opus 4.6 reasoning datasets floating around on Hugging Face. Figured I'd share the result! The main thing that makes this useful is that it's strict - every row, every assistant turn has reasoning_content populated. No partial coverage, no rows where reasoning just happens to be on the last turn. If a multi-turn… See the full description on the dataset page: https://huggingface.co/datasets/Gryphe/Opus-4.6-Reasoning-24k.

<iframe src="https://silytics.si/embed/item/a8860e26-35bc-4e91-870a-297e0e7608eb" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 17 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.