Every reply in a long 𝗛𝗲𝗿𝗺𝗲𝘀 Agent chat sends the entire chat to the model again. 🤯
This is a dev post classified by AGK Intelligence as AI dev tools (a tutorial), kept by the Dev Radar because it carries real work, not commentary.
Every reply in a long 𝗛𝗲𝗿𝗺𝗲𝘀 Agent chat sends the entire chat to the model again. 🤯 That's why reply 200 feels slower than reply 2. These 5 settings keep it short. 5 𝗛𝗲𝗿𝗺𝗲𝘀 settings for when long chats get slow👇 1. Finished one part of the job? → /compress here 2 ▸ Summarizes everything before your last 2 exchanges ▸ Keeps those 2 word for word, so the thread survives Needs: nothing. Works in the terminal and in Telegram. 2. Big 1M-window model? → compression.threshold_tokens 256000 ▸ By default it waits until ~500K tokens to compact ▸ This compacts at 256K, so each reply re-reads less Nee
Posted by Hermes Super-Intel (4.8k followers) 2 h ago · 7 likes · 725 views · view the original post on X. Kept by the Dev Radar as AI dev tools.
More dev work like this
- 想系统学 AI Agent，这本开源书值得从头啃一遍。 — @bkdgiffug
- I decompiled probably my favorite game ever: Battlefield 1942 and I put it on the web… — @levelsio
- Free 90-min masterclass on Claude agents from an Anthropic engineer (better than most… — @0x0SojalSec
- Bring your questions about PyTorch 2.15 to our live Q&A this Thursday, November 5, at 11… — @PyTorch
- “The Matrix is everywhere. It is all around us.” — Morpheus — @mishig25
- Meet Apriel-Voice-27B-preview: a new audio-text-to-text model from ServiceNow-AI. It… — @HuggingModels
- Stop rebuilding memory and tool wiring for every agent. — @DanKornas
- 😱卧槽！万物皆可逆向！Rust神器craft-anything炸场，看懂任何闭源软件，AI直接帮你重造一个自己的版本！ — @NFT_Chen
Every post is read and classified by AGK Intelligence: what it is, which market it belongs to, and whether the link is a real tool. 31.4k posts from 5k X accounts over the last 21 days, 9.1k tools, 12 markets. New posts are collected several times an hour and the full ranking is recomputed every hour — last update 2026-11-02 22:12 UTC. Full method.