A small Hermes Agent test shows faster GPT 5.5 Instant replies
Inside , the same short request was answered faster by than by . The request was very simple: tell a joke.
answered in about 3 seconds, while took about 9 seconds. That made a difference of about 6 seconds.
One short delay may not matter much for a single question, but repeated delays can matter when an agent handles many requests. also promoted its Agent OS, Hermes , support, and courses for turning faster models into systems.
Key points
- and were compared inside with the same request.
- answered in about 3 seconds.
- answered in about 9 seconds.
- The reported gap was about 6 seconds on one simple joke request.
- The result is useful as a speed signal, but too small to prove real .