Comparing GPT generations on alt-history scenarios: gpt-6 is 'another level'
prd_008 · x · 2026-11-01
A user shares how different GPT generations handle alternate-history scenario exploration:
- 4o: amusing scenarios but superficial and unbelievable
- o3: found good sources and reasoned through them, but prose and tables sucked
- gpt-5: only a small improvement; the in-product Q&A experience was the bigger difference
- gpt-6 in chat: better conversational style, improved reasoning and synthesis, plus visual intelligence — the author spent 2 hours on a flight going down rabbit holes
Screenshot shows gpt-6 rewriting history with the Marathas winning the Third Battle of Panipat, building "The Maratha Century: 1761–2026" with sourced narrative and historical maps.
More from Models
- GPT-6.1 Terra builds 3-level game in 6.5 min; DeepSeek V4.1 Flash slower but 36% cheaper — latent_lurker8587 · 2026-11-01
- YC-backed RightNow launches streaming decision model with 30ms text decisions — iamrobotbear · 2026-11-01
- Andrew Wilson: LLMs are outrageously inefficient for most basic tasks — andrewgwils · 2026-11-01
- Reddit Users Prefer Qwen 27B's Chat Style, Calls Claude Opus's Tone 'CEO Attitude' — vogelvogelvogelvogel · 2026-11-01
- China Daily: Qwen-Powered Apple AI to Launch in China in December — nordicinst · 2026-11-01
- Google Student Deals: AI Plus Free for 12 Months, AI Pro 74% Off up to 4 Years — iliaghp · 2026-11-01