Opper

Context Reasoning

Context understanding and reasoning tasks test accurate answers grounded in provided context. This capability is essential for knowledge-base support bots, policy lookup systems, and internal knowledge Q&A applications. Models are evaluated on their ability to provide accurate answers that are properly grounded in the given context rather than hallucinating information.

Input Tokens

Output Tokens

Duration

Cost

Context

Input

Order all the cities named in the text by time of visiting them

Expected output

1.Cartagena (14 Jun 1995) 2.Leticia (mid-Jun 1995, the day after Cartagena) 3.Manchester (3 Oct 2002 entry) 4.Tangier (mid-Apr 2004, one week before 22 Apr) 5.Chefchaouen (mid-Apr 2004, just after Tangier) 6.Granada (22 Apr 2004) 7.Córdoba 8.Sevilla 9.Cádiz 10.Málaga 11.Jaén 12.Toledo 13.Madrid 14.Ávila 15.Salamanca 16.Zamora 17.San Sebastián 18.Bilbao (8 May 2004)

Model output

Ordered list of cities by time visited (earliest → latest):

1. Cartagena — June 14, 1995 (current entry)
2. Leticia — planned "tomorrow" after June 14, 1995 (intended next-day travel)
3. Manchester — October 3, 2002 (entry location)
4. Tangier — the week before April 22, 2004 (crossed from Tangier)
5. Chefchaouen — spent four nights immediately before Granada (mid-April 2004)
6. Granada — April 22, 2004
7. Córdoba — after Granada (first stop in the Spain sequence)
8. Sevilla — after Córdoba
9. Cádiz — after Sevilla
10. Málaga — after Cádiz
11. Jaén — after Málaga
12. Toledo — after Jaén
13. Madrid — after Toledo
14. Ávila — after Madrid
15. Salamanca — after Ávila
16. Zamora — after Salamanca
17. San Sebastián — after Zamora
18. Bilbao — May 8, 2004 (current entry)

Mentioned but not clearly visited by the narrator: Valencia (Tomás "ended up in Valencia without me")