A test of an opencode plugin that tries to cut AI coding costs
TokenWarden.ai is a plugin for that aims to reduce how much spend on . A compares it with other plugins to see whether they actually lower usage costs. AI results can change from run to run, so 60 tests were run and averaged instead of relying on one result.
The tests used Qwen 3.7 Max and a , Qwen 3.5 9B. was also tried, but it failed many tasks and was not useful for the comparison. Some runs showed a strange result where every plugin used more than three times as many credits instead of saving them.
The likely cause was the AI taking a different work path or getting stuck in a loop, especially because the tests did not use top western models. The is still unfinished, with more tests and broader coverage planned.
Key points
- TokenWarden.ai is an plugin for reducing bills.
- The used 60 test runs and averaged the results.
- The main models tested were Qwen 3.7 Max and the Qwen 3.5 9B.
- Some runs showed all plugins using over three times more credits, which points to unstable AI behavior.
- More tests are planned, so the results should be treated as early rather than final.