Same prompt: MiniMax M3.1 Flash 1h06m/$0.22 vs DeepSeek V4.1 Flash 1h28m/$0.54
On one task M3.1 Flash finished 22 minutes faster and 2.4x cheaper; DeepSeek burned 89.5M tokens but billed $0.54 thanks to near-total cache hits, so the author notes your bill depends on cache hit rate.
DEEPSEEK BURNED 11X MORE TOKENS THAN MINIMAX ON THE SAME PROMPT. IT STILL PAID 54 CENTS. THE CATCH IS THE CACHE. round 1. same prompt, same task, the two new flash models: > deepseek v4.1 flash: 1h 28m, 89.5M tokens, $0.54 > minimax m3.1 flash: 1h 06m, 8.1M tokens, $0.22 minimax won the round. 22 min faster, 2.4x cheaper. but look at that deepseek token count. at the normal input price, 89.5M tokens is $26.85. deepseek charges $0.006 per 1M on a cache hit. 89.5M × $0.006 = $0.54. the exact bill. so the agent spent almost the whole run rereading its own context from cache. the price list isn't your bill. your cache hit rate is: > keep the system prompt and file order stable > don't let the agent rewrite early context mid run > every cache miss bills at full price today deepseek open sourced deepinfer, the engine it says runs v4.1 flash at scale. the whale's kitchen is public now. the cheapest model is the one that remembers what it already read. one prompt, two models, real receipts. a new fight every day.
This post quotes another post: x.com/hamelhusain/status/2112588257104216294