A brief update with some numbers for hardware load-balanced mongrels. 4000 requests/second on 48 mongrels behind a hardware load balancer.
Recent articles
- GPT-5: Key characteristics, pricing and model card - 7th August 2025
- OpenAI's new open weight (Apache 2) models are really good - 5th August 2025
- ChatGPT agent's user-agent - 4th August 2025