Hermes Agent Integrates OpenAI Native Server-Side Compaction for GPT-5.6
Teknium · x · 2026-08-10
Hermes Agent highlighted a new setting for GPT-5.6: support for OpenAI's native Responses API server-side compaction.
Designed for long-running sessions, the feature offers several benefits:
- Leverages OpenAI to compact older context server-side first, improving long-session management
- Keeps an encrypted compaction checkpoint and replays it on future turns
- Reduces unnecessary transcript baggage
- Retains Hermes' standard local compression as a fallback
The feature is not blindly enabled across all providers; it is specifically gated to GPT-5.6 running via the direct OpenAI API or a ChatGPT Codex subscription.
More from coding & agent
- DIRECT: retrieval-based next-question suggestions for conversational agents — burkov · 2026-10-16
- Claude Code and Codex now autocomplete the user's next action — tokenbender · 2026-10-16
- Hermes Agent ships official Reverse Engineer Anything plugin with one-command setup — Teknium · 2026-10-16
- MLExpert: 35B MoE Agent Trained to Autonomously Handle Machine Learning Engineering — Shangheng Du · 2026-10-16
- Cowork-Gym: Alibaba Synthesizes 64.9K Agent Work Environments, Boosting Qwen by 8.7 Points — Anni Zou · 2026-10-16
- MMSearch-OPD: Joint RL and On-Policy Distillation for Long-Horizon Multimodal Search Agents — PekingUniversity · 2026-10-16