My cheap coding plan ran out in 20 days, and cost more per merged change than GPT-6.1 Sol
Short version: I paid ¥329 (about $46) a month for Xiaomi's MiMo Token Plan Pro and burned through it in 20 days. Per accepted change, MiMo came out at about $0.81 and GPT-6.1 Sol at $0.71. The price per change was close. The real gap was rework: Sol passed review on the first try 71% of the time, MiMo 16%. The setup I run a small dispatcher called agent-dispatch. Claude plans the work, writes…
My cheap coding plan ran out in just 20 days, costing more per merged change than GPT-6.1 Sol. The plan, called MiMo Token Plan Pro, cost ¥329 ($46) a month for 38 billion credits. The idea was to have one Chinese model do the bulk coding, using the Claude quota for planning and review. However, the plan ran out on November 3, leaving only 42.7 billion credits actually used.
MiMo took 87 orders and passed 14 on the first try, while GPT-6.1 Sol took 31 orders and passed 25. The cost per accepted change was about $0.81 for MiMo and $0.71 for Sol. The main difference was the rework: MiMo required 2.7 extra attempts per accepted order, adding roughly $0.54 in review costs. The real cost per accepted change was close, with MiMo at $0.81 and Sol at $0.71.
The Chinese models are cheap, but the real gap lies in rework and review costs. The plan turned out to be just a prepaid card, not a heavily subsidized deal. Unused credits expire, and the plan does not roll over to pay-as-you-go. A more accurate comparison should consider cost per accepted change, including retries, review costs, and wall-clock time, rather than just price per token.
Written by urgent.news from Dev.to's reporting — not their text. Machine-written — may contain errors; check the original before relying on it.
