To save money, I gave Opus 5.5 a cheap DeepSeek V4.1-Flash-1017 to write the code... but the accuracy dropped and the bill went up 🫠
Opus 5.5 alone: 81% of tasks, $1.20 per task
Opus + DeepSeek V4.1-Flash-1017: 67%, $1.29 per task
DeepSeek’s Super-agent post tested scheduling on simulated coding (not real tasks). I tested 7 team setups on 12 real, hard ML coding tasks.
> Telling Opus “your tokens cost 20x more” instead of banning it from coding: the worker wrote 94% of the code, same 67%, and the cost per task dropped to $0.71
> Teams aren’t cheap: Opus re-caches its whole context after every long worker call, and that costs more than the worker itself
> 71% of team failures are tasks the worker also fails alone
Thread ↓
Science
Claude Opus plus a cheaper model raised cost and dropped accuracy
Ibragim, a researcher at Nebius AI, compared Opus 5.5 alone and paired with DeepSeek V4.1-Flash-1017 on 12 ML coding tasks. Accuracy fell from 81% to 67% and cost per task rose from $1.20 to $1.29; telling Opus its tokens cost 20× more cut cost to $0.71.
467likes
37reposts
457saves
71Kviews
- Made with
- Opus 5.5, DeepSeek V4.1-Flash-1017
- Role
- Researcher at Nebius AI
- Posted
