English(EN) @ ppulfer ML (machine learning) is here to stay, especially running "small" models locally, trained on high quality, narrow context data by a ML expert No one n

本地机器学习模型比昂贵、不可持续的大型语言模型更受欢迎

作者 PulseAugur 编辑部 · [1 个来源] · 2026-05-14 20:39

在本地运行小型机器学习模型，利用专业数据，被认为是比托管在远程服务器上的大型语言模型更可持续、更具成本效益的替代方案。该论点认为，基于云的大型语言模型的真实成本，包括硬件、能源消耗和利润率，使其成为一项不合理的投资，且没有明确的盈利途径。这种观点提倡使用本地化、专家训练的模型，而不是当前大规模、集中式人工智能的趋势。 AI

影响提倡使用本地化、专家训练的模型，而不是大规模、集中式人工智能，暗示了机器学习资源部署方式的转变。

排序理由该集群包含一篇讨论本地机器学习模型与基于云的大型语言模型优点的观点文章。

在 Mastodon — fosstodon.org 阅读 →

AI 生成摘要 · Google Gemini · 来自 1 个来源。我们如何撰写摘要 →

报道来源 [1]

Mastodon — fosstodon.org TIER_1 English(EN) · [email protected] · 2026-05-14 20:39

@ ppulfer ML (machine learning) is here to stay, especially running "small" models locally, trained on high quality, narrow context data by a ML expert No one n

@ ppulfer ML (machine learning) is here to stay, especially running "small" models locally, trained on high quality, narrow context data by a ML expert No one needs LLM running on someone else's overheating unsustainable hardware No one can afford cloud # LLM : objectively there …

链接 mastodon.ie/@ppulfer posthat.ca/@Solarinas

报道来源 [1]

@ ppulfer ML (machine learning) is here to stay, especially running "small" models locally, trained on high quality, narrow context data by a ML expert No one n

相关实体

相关话题