Comprehensive benchmark testing deep factual knowledge, nuance, chronological accuracy, and resistance to historical misinformation across Swedish history from the Kalmar Union to the 20th century.
Scorer: ClosedQA
loading chart
| # | score | model | cost price | ttft | total tok |
|---|---|---|---|---|---|
| 1 | 100% | claude-opus-5 | $0.08535 $5.00 / $25.00 per 1M | 1.42s | 3,471.7 |
| 2 | 80% | openai/gpt-5.6-terra-pro | $0.02729 $1.00 / $6.00 per 1M | 30.26s | 8,650.7 |
| 3 | 40% | mistral-large-2512 | $0.00198 $0.50 / $1.50 per 1M | 1.09s | 1,352.1 |