Ways to think about token pricing
mercutio2
40 points
15 comments
July 09, 2026
Related Discussions
Found 5 related stories in 104.5ms across 9,716 title embeddings via pgvector HNSW
- Claude Code pricing: same tokens, same model, up to 40x the price jakozaur · 30 pts · August 11, 2026 · 52% similar
- Show HN: Frugal Tokens – explore costs and usage across coding agents dpc94 · 29 pts · August 19, 2026 · 51% similar
- Tokens too cheap to meter teoruiz · 273 pts · September 23, 2026 · 51% similar
- Is the industry ready for tokens-constrained work? fullautomation · 40 pts · August 16, 2026 · 49% similar
- TokenTown: A visual way to understand how LLMs work laurentiurad · 71 pts · July 29, 2026 · 48% similar
Discussion Highlights (2 comments)
nekusar
1. Slot machine You put in minutes of time combined with token costs. You pull the lever and HOPE something good comes back. And it gets tantalizingly close, but for 100% gotta pull the arm again!
mips_avatar
I know there's a lot of reasons to think that everyone will just use AI inference in the cloud, but I think if everyone had access to a dgx gb400 class machine with 512gb of hbm4 vram and 1tb of lpddr8x a lot of people are going to be running finetunes of models locally. Like the dgx gb300 is $94k now, but i bet this class of machine will come down to $20kish in the next 2-3 years.