Summary
- Answers 59 of 80 verification questions correctly (73.8%), placing #18 of 72 models tested on the current suite.
- Generates 122 output tokens per second, faster than 62% of the models we measure.
- Starts responding after 2.4 s (median time to first token), including thinking time.
- Costs $0.0013 per question as billed, #2 cheapest of 31 models priced over $1 per 1M tokens.