Hands-on: Opus 5.5 high beats GPT-6 astra xhigh on real Pagespeed optimization
mazzaTalk · x · 2026-09-27
The author tested frontier models on optimizing a website's Pagespeed and found GPT-6 astra xhigh failed to beat Opus 5.5 high. They also note that Codex is now the one being called underwhelming.
A first-hand horizontal comparison of frontier models on a real-world task.
More from Models
- ChatGPT subscription tiers relabeled "Standard" and "More", usage multipliers removed — ssh4net · 2026-09-27
- Claude Opus 5.5 makes its own 20-page sketchbook: handwriting, doodles, and piano music — CurieuxExplorer · 2026-09-27
- OpenAI strips 5x/10x/20x usage multipliers from plan upgrade UI, leaving vague wording — ssh4net · 2026-09-27
- Puppy Kill Bench: most models refuse, GPT6-Luna just executes the kill tool — MetroidsSuffering · 2026-09-27
- Ethan Mollick: Opus 4.7-5 lost the 'Claude feel', Opus 5.5 brings it back — emollick · 2026-09-27
- Martin Casado recommends the best talk on in-context learning, a first-principles view of LLMs — AccBalanced · 2026-09-27