tokenjam 0.5.2 n=5 tasks · k=1 sample(s) · openai:gpt-4o → openai:gpt-4o-mini
We ran the same 5 judged task(s) through both models and graded every answer with the same automated judge. The cheaper candidate (gpt-4o-mini) cost 93% less than gpt-4o, and scored 20 points higher on this suite — 40% of tasks passed before, 60% after.
Too few tasks were run to be statistically sure either way. Run more before deciding.
| Task | Original | Cheaper model | Why — the judge’s reason |
|---|---|---|---|
| Refund policy | fail | fail | The actual output provides a general explanation of what a refund window is and graded 0.30 (needs 0.50) |
| Capital | pass | pass | The actual output matches the expected output exactly in terms of factual inform graded 1.00 (needs 0.50) |
| Retry summary | pass | pass | The actual output accurately reflects the factual information of the expected ou graded 0.90 (needs 0.50) |
| Shipping | fail | fail | The actual output provides a detailed breakdown of standard shipping times based graded 0.30 (needs 0.50) |
| Define llm | fail | pass | The actual output provides a broader description of a large language model, focu graded 0.50 (needs 0.50) |
McNemar’s test asks whether the difference between the two models is bigger than chance: a p-value above 0.05 means the change is not statistically significant. The 95% CI on the pass-rate delta is the range the true difference is likely in — if it crosses zero, the direction isn’t certain.