Try these in your first Opus 5.5 session:
โ Hand over a whole task. Define "done" and when to check in.
โ Drop "think carefully". It always thinks first.
โ After a long run, check what it needs to go further.
Our playbook: https://t.co/h4Vz9BIl0z
Please welcome GPT-6 Sol and GPT-6 Luna to the GPT-6 universe.
GPT-6 Sol and Luna build on the advances behind GPT-6 Astra, bringing much of its strengths into faster and more affordable models to support work at scale.
Weโve also made caching and inference more efficient, and weโre passing the savings directly to you: 50% lower API prices for Sol and Luna compared with GPTโ5.6 promotional pricing.
This is Opus 5.5 I find the quality of the game to be a bit better but the real difference is in price the same game cost $16 with Opus 5 and just $6 with Opus 5.5!
it took about the same time to finish on both models.
Introducing Claude Opus 5.5, the first model in our new Claude 5.5 family.
It performs at the level of Claude Fable 5.1 for most tasks, and costs 40% less to run than Opus 5.
I know making a 2D game isnโt the best test, but I did a quick comparison between the new Grok 4.7 and Opus 5. I find them pretty close, but I feel like Opus still did a better job with the sound effects, animations, and overall feel of the game. Grok also messed up the background a bit, but itโs still way cheaper.
Grok took 41 minutes and cost $1.57. Opus took 35 minutes and cost $16.61.
Grok left, Opus right.