Teknium Claims Big Token Efficiency Gains in Hermes; User Reports 89% Savings vs Codex
Teknium · x · 2026-09-07
Nous Research's Teknium says Hermes made huge token efficiency improvements over the past two weeks, inviting users to try codex subagents in Hermes Agent. A quoted user reports the same workload that burned 2.5 Astra Medium resets in the ChatGPT Codex app in half a day used only 11% of quota all day with Hermes ULTRA and FAST modes — the first hard evidence of "less is far more" in Hermes.
More from coding & agent
- Dev demos AI models handling 3D objects and animations surprisingly well — chongdashu · 2026-10-23
- Jason Kneen burns 104.56B tokens and $24.7K in 304 days of AI coding on Codex and Claude Code — jasonkneen · 2026-10-23
- Typosquatted Blender MCP Repo Hit 958 Stars in Days and Silently Ran Malware — Promptmethus · 2026-10-23
- Benchmarking 56 fresh dependency bugs: MCP routing lifts Claude Code accuracy 68% to 91% — Available_Side_7134 · 2026-10-23
- Swapping only the eval harness moves results 30%: designing agent evals like a scientist — AI Engineer · 2026-10-23
- Med student builds a whole-library RAG with Cognee; single-book ingestion still freezes his machine — shrike293103 · 2026-10-23