7 tweaks cut this agent's token cost by up to 95%: 40,000 to 19,200 tokens without changing models
tomcrawshaw01 · x · 2026-11-04
Tom Crawshaw shares 7 token-cutting changes that made his Hermes Agent up to 95% cheaper to run. A free audit plugin dropped overhead from 40,000 to 19,200 tokens before changing a single model. He highlights Ponytail, a plugin that trims fat while the agent writes code. Full setup walkthrough is in his article plus a YouTube video with the Token Optimizer tool.
More from coding & agent
- Apple's new Siri AI dev framework officially adopts Markdown-based SKILL.md agent files — rudrank · 2026-11-04
- Baseten ships Grid gateway, Sandboxes and Loops serverless training this week — baseten · 2026-11-04
- Sentry's AI Agent goes head-to-head with Semgrep bot in a real PR review — zeeg · 2026-11-04
- One Bland agent replaces 300 dealership agents, handling 12,000+ calls a day — ycombinator · 2026-11-04
- Snowflake open-sources Arctic Platform, a modular post-training framework for LLMs — StasBekman · 2026-11-04
- Sentry Engineer Shares Bots Arguing on a PR: Security Bot Flagged, AI Bot Refused and Resolved — zeeg · 2026-11-04