Heavy OpenAI user does the math: subscription changes are equilibrium, not a rug pull
fluffy_serval · reddit · 2026-09-29
A power user argues OpenAI's subscription changes aren't a rug pull: the old $200/mo Pro plan delivered multiples of equivalent API value on increasingly better models — subsidized compute that was never permanent.
Key points:
- The token unit has become hopelessly opaque, baking in model choice, reasoning depth, caching, retries and agent behavior; forcing users into "token math" is bad product.
- He'd rather have an "automatic transmission for models": OpenAI routes planning and execution, eats the cost of bad choices, in exchange for a pricier but predictable Pro subscription.
- He runs the latest models on a 600,000+ line codebase, would pay even double the price, but wants the models' token-burning testing habit fixed immediately.
Verdict: it's market equilibrium and a needed end to token opaqueness — and OpenAI has goodwill to deliver on.
More from Models
- Google Puts Gemini 3.5 Flash Inside Air-Gapped Customer Sites in Sovereign AI Push — shashib · 2026-11-15
- Reddit user reports Gemini Live replying in a clone of their own voice — wayfahrter · 2026-11-15
- Google's AI suddenly turned argumentative and stopped apologizing, users report — ImNotForJerks · 2026-11-15
- After a week with Carbon: smarter than Opus 5.5, but the best workflow pairs both — rudrank · 2026-11-15
- Meituan's 1.6T-parameter LongCat-2.5 agent model trained without Nvidia GPUs — The Decoder · 2026-11-15
- User builds entire single-file project with Gemini — no libraries, no assets — fofrAI · 2026-11-15