Stories by npmipg

OS 12B model that beats Claude 4 Sonnet at video captioning and is 17x cheaper

GPU-rich labs have won: What's left for the rest of us is distillation

The Economics of Hosting Open Source Models

Smart Routing Saved Exa 90% on LLM Costs

The Cheapest LLM Call Is the One You Don't Await

1T Tokens for Sale

Late Chunking vs. Contextual Retrieval: The Math Behind RAG's Context Problem

Show HN: Open Source 12ft.io Alternative With Free AI Summaries

Show HN: SMRY, AI-enhanced open source 12ft.io and removepaywall.com alternative

Show HN: Slang Thesaurus – An AI-Powered Slang Thesaurus Trained on UrbanDict