{"slug": "inkling-mini-26b-a3b", "title": "Inkling Mini - 26B A3B", "summary": "Thinking Machines released Inkling-2-Mini, a 26B-parameter image-text-to-text model on Hugging Face that the release claims matches or beats Qwen 3.8 27B on most benchmarks. The model uses a looping architecture that loops 3 times per token. The release gives no further benchmark figures or technical details beyond the parameter count and loop count.", "body_md": "huggingface.co thinkingmachines/Inkling-2-Mini · Hugging Face Image-Text-to-Text model with 26B parameters. Download, explore files and run it on Hugging Face. As good or better than Qwen 3.8 27B on most benchmarks. Looping architecture - it loops a total of 3 times per token.\n\nI too am interested in seeing this optimized for 2xr9700s.", "url": "https://wpnews.pro/news/inkling-mini-26b-a3b", "canonical_source": "https://forum.level1techs.com/t/inkling-mini-26b-a3b/260913#post_2", "published_at": "2026-11-21 03:38:02+00:00", "updated_at": "2026-11-21 03:42:14.632682+00:00", "lang": "en", "topics": ["large-language-models", "machine-learning", "ai-research"], "entities": ["Thinking Machines", "Inkling-2-Mini", "Hugging Face", "Qwen 3.8 27B"], "also_reported_by": [], "takeaway": "Inkling-2-Mini is a 26B-parameter image-text-to-text model that loops its architecture 3 times per token and is claimed to match or beat Qwen 3.8 27B on most benchmarks.", "recommendation": null, "alternates": {"html": "https://wpnews.pro/news/inkling-mini-26b-a3b", "markdown": "https://wpnews.pro/news/inkling-mini-26b-a3b.md", "text": "https://wpnews.pro/news/inkling-mini-26b-a3b.txt", "jsonld": "https://wpnews.pro/news/inkling-mini-26b-a3b.jsonld"}}