←新闻详情

Testing Catalog更新于 10月21日 22:46

Paper Instruments 发布开源模型 Ultramarine 1 Flash

Paper Instruments has released Ultramarine 1 Flash, its new open model made for document-heavy workloads of banks, law firms, and consultancies.

AI 导读

Paper Instruments 发布开源模型 Ultramarine 1 Flash,面向银行、律所和咨询机构的文档密集型工作负载,可生成 Excel 模型、幻灯片或 Word 备忘录。该模型由 GLM-5.3 Flash 后训练而来,在其内部客户交付物基准上得分 59.3、每任务成本 0.13 美元,对比 Claude Fable 5 的 57.8 分和 8.19 美元。引用帖称该模型由 5 人团队后训练、总投入约 30 万美元,现已上线 API,权重将在两周后发布。

推荐理由

AI 评分 71

原文给出内部基准上的分数与单任务成本对比,读者可据此判断文档类工作负载的性价比变化。

主题

标签

引用

Daanish Khazi

@bertgodel

We are releasing Ultramarine 1 Flash, a frontier-class model built for knowledge work. It is open-source and achieves similar performance to Fable 5, while being 60x cheaper. It was post-trained by a team of 5 for ~$300k all-in. Available on API now, with weights releasing in 2 weeks.

查看原帖

正文

当前显示原文,中文全文译文尚未就绪。

Paper Instruments has released Ultramarine 1 Flash, its new open model made for document-heavy workloads of banks, law firms, and consultancies.

The company states the model can deliver Excel models, slide decks, or Word memos for cents per task.

> The model is a post-training of GLM-5.3 Flash.

> On Paper Instruments' internal benchmark for client-ready deliverables, it scored 59.3 at $0.13 per task, versus 57.8 at $8.19 for Claude Fable 5.

> Each training task was graded on its own rubric, since a diligence memo or a board deck has no single correct answer.

来源

Testing Catalog