ChatGPT Plus Users Ask Which Models Check Proofs

Hot take: this is less a product announcement and more a reality check on how far model reasoning still has to go before users trust it on proof verification.

• –The post is anchored on ChatGPT Plus, so ChatGPT is the reference product being discussed.
• –The thread has almost no hard evidence; the only concrete pointer is MathArena, which suggests users want benchmark-backed comparisons.
• –Community sentiment in the comments is skeptical of Gemini Deep Think and generally favors OpenAI's higher-tier reasoning offerings.
• –Because the discussion is about practical reliability on math proofs, it reads as a use-case judgment thread rather than a launch or update.