An OpenAI benchmark paper suggests that the Pro tier of GPT-5.6 could ship in three variants. That would be the first major change to ChatGPT Pro's structure since the plan launched.
OpenAI officially unveiled the GPT-5.6 generation in late June, splitting it into three models. Sol handles the hardest tasks, Terra targets high-volume business workloads, and Luna covers faster, cheaper everyday queries. Pro variants weren't part of the announcement.
Now a new OpenAI paper on a genomics benchmark reveals Pro models for the first time. The results table includes rows for "GPT-5.6 Luna Pro," "Terra Pro," and "Sol Pro," each labeled as "Pro (Extended)" runs.
Pro is no longer just one top-tier model
In the benchmark, Sol Pro hits a pass rate of 31.5 percent, making it the strongest of all 60 tested models. It beats the standard Sol at 28.7 percent and the best non-GPT score, Claude Opus 4.8 at 16.0 percent. The pass rate measures how often a model completes the full multi-step analysis without errors and arrives at the correct final answer.
