惯性聚合 高效追踪和阅读你感兴趣的博客、新闻、科技资讯
阅读原文 在惯性聚合中打开

推荐订阅源

腾讯CDC
OSCHINA 社区最新新闻
OSCHINA 社区最新新闻
博客园 - 叶小钗
人人都是产品经理
人人都是产品经理
博客园 - 聂微东
The Cloudflare Blog
爱范儿
爱范儿
阮一峰的网络日志
阮一峰的网络日志
WordPress大学
WordPress大学
小众软件
小众软件
博客园 - 三生石上(FineUI控件)
Last Week in AI
Last Week in AI
Jina AI
Jina AI
V
V2EX
罗磊的独立博客
V
Visual Studio Blog
A
About on SuperTechFans
IT之家
IT之家
P
Proofpoint News Feed
B
Blog
博客园 - Franky
Blog — PlanetScale
Blog — PlanetScale
Google DeepMind News
Google DeepMind News
Y
Y Combinator Blog

Hacker News: Ask HN

The New Window Delete ChatGPT Atlas Spyware Tell HN: Qwen Free Tier Is Discontinued Ask HN: SeedLegals Partnerships in London, worth it? Ask HN: How to highlight talent from untraditional backgrounds? Ask HN: We dont need a programming language now? Durable Object alarm loop: $34k in 8 days, zero users, no platform warning What if Time at the subatomic level has multiple arrows? How to add MidnightBSD Key to UEFI Secure Boot DBX? (Revoked and Forbidden Keys) Ask HN: What's your experience working at xAI as an AI tutor? Any engineers here with experience of clinical data standards? Ask HN: Who is using OpenClaw? Agent Skills for Software Test Automation Ask HN: Who needs contributors? Claude Code is thinking too much Ask HN: What Is the Big-O Order of a Jigsaw Puzzle? Ask HN: Stepping into a new role as a Senior, mentoring dos and dont's? Founder from Zurich heading to SF and Austin for the first time Hacker News No Manual Screenshots: I Built a Scalable Screenshot API Using Cloud Playwright Ask HN: Thought experiment: AGI giving us answers we don't like? Ask HN: I quit my job over weaponized robots to start my own venture 1% Vacancy, 81% Preleased: Where Midmarket Compute Deploys in 2026 Ask HN: Preferred pricing model for sound effects libraries? Copy of the email I sent to my undergraduate professors on Nov 30, 2025 Model API Performance | Hacker News Ask HN: Are open-weight LLMs the new offline encyclopedias? Valgrind 3.27 RC1 is out Claude Code OAuth down for >12 hours Ask HN: What's Better?–Tauri or Electron?
Changes that cut our LLM pipeline costs more than model-s...
Abbas_Maka · 2026-06-20 · via Hacker News: Ask HN

I have been building multiple LLM systems and for our Organization biggest cost savings weren't from prompt-wordsmithing or model switchings. Sharing useful to anyone watching their token bill :

1) JSON → TOON for structured output: JSON was not made for LLMs. well you can implement your own verison that fits for your needs that reduce tokens usage but what worked for us was TOON. TOON cut output our tokens by ~30% same information, way less syntax tax.

2) Full markdown/HTML → condensed markdown: Using markdown for writing your prompts, getting intermediate results or communication between your Agents eats a lot of tokens. We swithced to condesed markdown and short system prompts that replicate Caveman. this alone cut just on input token costs ~50% on calls that pass prior context forward which can be implemented between Agent Calls.

3) Long Do/Don't instruction lists → 2-3 multi-shot examples: Counterintuitive one - replacing a large lists of DO's and Don'ts for agents rules don't help. rather couple of concrete examples that convers major and all cases actually improved output quality more reliably and it's usually fewer tokens once the instruction list gets long enough to cover real edge cases.

I have seen most people on this sub reddit talk about using open-source or cheaper models. Like we were spending thousands of dollar's but this all changes alone helped reduce cost by 60%.

edit: Open to Discussion, anyone whether something similar would help their setup.