14× faster embeddings: how we rebuilt the ONNX path in Manticore (manticoresearch.com) | HN Screenshots

YHacker Newsscreenshots

Screenshot of manticoresearch.com: 14× faster embeddings: how we rebuilt the ONNX path in Manticore

First screen, captured 3 months ago.

Looks similar

Screenshot of neuml.github.io: Txtai: Open-source vector search and RAG for minimalists

Txtai: Open-source vector search and RAG for minimalists (neuml.github.io)

220 points by dmezzetti 2 years ago | 48 comments

Screenshot of manticoresearch.com: Better Vector Search for Long Documents: Chunking Inside Manticore Search

Better Vector Search for Long Documents: Chunking Inside Manticore Search (manticoresearch.com)

71 points by GloriaVinogrado 1 month ago | 11 comments

Screenshot of meshtastic.org: Meshtastic: An open source, off-grid, decentralized, mesh network

Meshtastic: An open source, off-grid, decentralized, mesh network (meshtastic.org)

267 points by 882542F3884314B 2 years ago | 56 comments

Screenshot of deepseek.com: DeepSeek v2.5 – open-source LLM comparable to GPT-4, but 95% less expensive

DeepSeek v2.5 – open-source LLM comparable to GPT-4, but 95% less expensive (deepseek.com)

149 points by jchook 2 years ago | 41 comments

Screenshot of digital-naturalism-laboratories.github.io: Mothbox 4.0

Mothbox 4.0 (digital-naturalism-laboratories.github.io)

258 points by bookofjoe 2 years ago | 38 comments

Screenshot of help.kagi.com: Kagi is 80% faster and 70% less carbon intensive than Google

Kagi is 80% faster and 70% less carbon intensive than Google (help.kagi.com)

36 points by JumpCrisscross 3 years ago | 20 comments