{"slug": "inception-launches-mercury-3-at-1068-tokens-per-second", "title": "Inception launches Mercury 3 at 1,068 tokens per second", "summary": "Inception launched Mercury 3 on October 19th, a diffusion-based language model that CEO and co-founder Stefano Ermon says delivers 1,068 tokens per second, a 1-million-token context window and benchmark gains over rival models, with launch pricing at half off. Inception positions Mercury 3 as a faster alternative to conventional systems that generate text one token at a time, and the company is testing whether diffusion can scale into general agent and business workloads while keeping its speed advantage. The published results and partner examples are company-reported, so adoption will depend on how the model performs in developers' own tasks.", "body_md": "# Inception launches Mercury 3 at 1,068 tokens per second\n\n**Stefano Ermon's diffusion model claims a 1-million-token context window and benchmark gains over rival models, with launch pricing half off.**\n\n        By [Ryan Merket](https://runtimewire.com/author/ryan-merket)\n        · Published \n\nPrimary source: [X](https://x.com/StefanoErmon/status/2112222911218291051)\n\n## Why it matters\n\nMercury 3 tests whether diffusion can scale into general agent and business workloads while keeping its speed advantage. Its published results and partner examples are company-reported, so adoption will depend on how the model performs in developers' own tasks.\n\n[Inception](https://www.inceptionlabs.ai/blog/introducing-mercury-3) launched Mercury 3 on October 19th, pitching its diffusion-based language model as a faster alternative to conventional systems that generate text one token at a time. CEO and co-founder [Stefano Ermon (@StefanoErmon)](https://x.com/StefanoErmon) says it delivers more than 1,000 tokens per second while improving on the company's earlier models.\n\nWhen you purchase through links in our articles, [we may earn a small commission](https://runtimewire.com/affiliate-monetization-standards). This doesn’t affect our editorial independence.", "url": "https://wpnews.pro/news/inception-launches-mercury-3-at-1068-tokens-per-second", "canonical_source": "https://runtimewire.com/article/inception-mercury-3-diffusion-llm-launch", "published_at": "2026-10-19 17:06:00+00:00", "updated_at": "2026-10-19 20:15:10.242016+00:00", "lang": "en", "topics": ["large-language-models", "generative-ai", "ai-research", "ai-products", "artificial-intelligence"], "entities": ["Inception", "Mercury 3", "Stefano Ermon"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/inception-launches-mercury-3-at-1068-tokens-per-second", "markdown": "https://wpnews.pro/news/inception-launches-mercury-3-at-1068-tokens-per-second.md", "text": "https://wpnews.pro/news/inception-launches-mercury-3-at-1068-tokens-per-second.txt", "jsonld": "https://wpnews.pro/news/inception-launches-mercury-3-at-1068-tokens-per-second.jsonld"}}