Summary
- Answers 1 of 80 verification questions correctly (1.2%), placing #72 of 72 models tested on the current suite.
- Generates 101 output tokens per second, faster than 49% of the models we measure.
- Starts responding after 0.5 s (median time to first token), including thinking time.
- Costs $0.0008 per question as billed, #11 cheapest of 12 models priced under $0.15 per 1M tokens.