https://blog.margrop.net/en/tag/token-compression/ Recent content in Token Compression on Margrop Blog Hugo en-US Sat, 20 Jun 2026 12:30:00 +0800 https://blog.margrop.net/en/post/headroom-newapi-openclaw-hermesagent-token-compression-guide/ Sat, 20 Jun 2026 12:30:00 +0800 https://blog.margrop.net/en/post/headroom-newapi-openclaw-hermesagent-token-compression-guide/ <blockquote> <p><strong>TL;DR</strong></p> <p>I did not replace NewAPI, and I did not point OpenClaw or HermesAgent at an unproven gateway. The actual design is simpler: place Headroom in front of NewAPI, then move only the already validated OpenAI-compatible providers to <code>http://&lt;headroom-host&gt;:8787/v1</code>. The original NewAPI endpoint stays available. Long agent context now goes through Headroom first, gets compressed, then continues to NewAPI for the same routing and model management as before.</p> <p>The rule that kept the rollout boring was: <strong>test first, edit second; migrate only OpenAI-compatible providers that pass; leave non-OpenAI fallbacks alone.</strong> This post is both a write-up and a runbook you can hand to an Agent or follow manually.</p> </blockquote>