Kyotofin open-sourced a tax-doc classifier built on Jev.
This is a AI post classified by Jev as RAG & memory (a tool drop), kept by the AI Radar because it carries real work, not commentary.
Kyotofin open-sourced a tax-doc classifier built on Jev. It classifies 100% of their tax-document dataset at about $0.001/page. Kyotofin says it’s 34× cheaper and 6× faster than their previous LLM-based pipeline, used to process thousands of documents. GitHub: http://github.com/kyotofin/tax-doc-classifier How far can specialized classifiers replace LLM pipelines?
Posted by David Arnal (1.5k followers) 2 h ago · 1 likes · 48 views · view the original post on X. Kept by the AI Radar as RAG & memory.
More AI work like this
- Nautilo agents use memory like memory competition champs. They never forget and they… — @Dan_Jeffries1
- Use ChatGPT Work to ground analysis in your semantic layer… — @cxocommunity
- we've evaluated 100+ models on doc parsing 🔥 — @jerryjliu0
- hello beautiful sir/mam, — @DhravyaShah
- Meet Vela-1.0-Omni-Mini: a feature-extraction model that bridges images, text, and audio… — @HuggingModels
- You guys spent 281 Million token in the last 24 hours on my jev demo, costing me around… — @iam_zachi
- Recently, Datalab released OmniExtractBench, a benchmark that combines several… — @jerryjliu0
- Long AI stories usually lose the plot by chapter nine: who knew the secret, who stopped… — @delx369
Every post is read and classified by Jev (TypeSafe): what it is, which market it belongs to, and whether the link is a real tool. 31.3k posts from 4.7k X accounts over the last 14 days, 1.3k tools, 19 markets. Collected every 5 minutes, fully re-ranked every hour — last update 2026-09-19 20:46 UTC. Full method.