gpt-5.6-sol Benchmark and Quota Issues

jxnlco · x · 2026-07-12

[Repost] A hands-on article reports that the author burned through over $200,000 worth of tokens using gpt-5.6-sol, concluding it's a 'very strong model.' However, the $200 Codex Pro subscription hits its cap easily; while OpenAI has been generous in resetting quotas, the limitation issue remains.

Related event: Testing GPT-5.6-sol Burns Over $200K in Tokens(3 posts)→

Original post →

More from Models

Models channel →