{"id":7489,"date":"2024-04-09T10:25:34","date_gmt":"2024-04-09T02:25:34","guid":{"rendered":"https:\/\/www.1ai.net\/?p=7489"},"modified":"2024-04-09T10:25:34","modified_gmt":"2024-04-09T02:25:34","slug":"stability-ai%e5%8f%91%e5%b8%83stable-lm2%e6%a8%a1%e5%9e%8b%e6%9b%b4%e6%96%b0%ef%bc%8c%e5%8f%82%e6%95%b0%e5%a2%9e%e8%87%b3120%e4%ba%bf","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/7489.html","title":{"rendered":"Stability AI releases Stable LM2 model update with 12 billion parameters"},"content":{"rendered":"

Stability<\/a> AI inup to date<\/span>Stable LM2 model<\/a>In the update, it was announced that the parameters will be increased to 12 billion to improve performance and accessibility. The update provides users with more powerful and accurate text content generation capabilities. Stability AI pointed out that the model performed well under specific benchmarks, surpassing larger models such as Llama270B.<\/p>\n

The new Stable LM212B models include a base version and a guided-tuned variant designed to enhance conversational skills in seven languages: English, Spanish, German, Italian, French, Portuguese and Dutch. The company said the new models are available for commercial use through Stability AI memberships, which is also one of the company's profit models.<\/p>\n

\u201cThe guided models are specifically designed to interact conversationally with users, and we\u2019ve worked hard to improve safety,\u201d said Carlos Riquelme, head of Stability AI\u2019s language team. The update comes less than a month after co-founder and CEO Emad Mostaque resigned, and the company\u2019s new interim co-CEOs Shan Shan Wong and Christian Laforte continue to roll out new model updates, showing that the company is still making steady progress.<\/p>\n

In addition to adding parameters to Stable LM2, Stability AI also improves on the previously released 1.6B version. Riquelme pointed out that smaller models such as the 1.6B Stable LM can be used to perform more specific and narrow tasks, while the larger 12B model has more features. However, he emphasized that the updated Stable LM21.6B model has improved its conversational style, is safer, and is better able to connect with other software tools.<\/p>\n

\"\"<\/p>\n

Different models will offer different trade-offs depending on the use case and available resources, such as response time, memory, and budget. \u201cWe believe there is no single model that works for every situation,\u201d Riquelme explained.optimal<\/span>model size, which is why we offer two rather different model sizes.\u201d<\/p>","protected":false},"excerpt":{"rendered":"

Stability AI, in its latest Stable MM2 model update, announced an increase in parameters to 12 billion to enhance performance and accessibility. The update provides users with a stronger and more precise capacity to generate text content. Stability AI noted that the model performed well under specific benchmark tests, exceeding larger models such as Llama270B. The new Stable MM212B model, which includes a base version and a guided variant, aims to enhance the conversational skills of seven languages (English, Spanish, German, Italian, French, Portuguese and Dutch). According to the company, these new models can be used commercially through the Stability AI membership, which is also one of its profitable models. S<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[2137,388],"collection":[],"class_list":["post-7489","post","type-post","status-publish","format-standard","hentry","category-news","tag-lm2","tag-stability"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/7489","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=7489"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/7489\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=7489"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=7489"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=7489"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=7489"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}