TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Show HN: ragapi – an API for using LLMs on your own data

12 点作者 riguer1超过 1 年前

3 条评论

riguer1超过 1 年前
Feel free to drop your email if you’re interested!<p>For context: As we talked with developers and product builders we noticed a common need for customising LLMs on their own data through fine-tuning (Retrieval Augmented Generation mainly, but some-times actual fine-tuning). Models like GPT, Claude and Llama2 have great reasoning capabilities but may not perform optimally for specific use cases where relevant information from knowledge sources is needed.<p>As we looked how this is done today it requires mastering a bunch of things from data retrieval, configuring vector DBs, data enrichement using embedding and ensuring things work not only for a few documents but for large amounts of data.<p>We&#x27;re building ragapi to manage all this heavy lifting so you can focus on building the rest of the (i.e use case related things).
AureliusMA超过 1 年前
Aren’t landing pages against hn rules?
评论 #37376667 未加载
skeptrune超过 1 年前
Hey, we have literally the same idea. Should get in touch.<p>nick@arguflow.gg