Evals

Added in 2026

• Arithmetic Fixed Cases · April 2026
• Czech Case Analysis · April 2026
• Heavy and Light Idiom Translation · April 2026
• Image Generation - Wineglass and Clock · April 2026
• Two Turn Constraint Shift · April 2026
• Czech Spring Poem — Three-Turn Revision · July 2026
• Šla Nanynka do zelí — Five Poets · July 2026
• Terra Blind Critique — Spring · July 2026
• Terra Blind Critique — Nanynka · July 2026

All categories

About these prompts

Spatial Cognition

• Recognition of Vertical Text
• Spatial Relationships in Language
• Visual Reasoning

Multilingual

• Multilingual: German
• Multilingual: Czech Metalanguage
• Multilingual: Czech Morphology
• Multilingual: Cultural Background
• Multilingual: Idiom Testing

Small Models

• Small Models: Knowledge & Instruction
• Small Models: Multilingual

Creative Writing

• Creative Writing: Wodehouse
• Creative Writing: Poetry

Long Context

• Long Context: Anachronisms
• Long Context: Retrieval

Coding

• Coding Capability & Reasoning

Linguistics

• Metalanguage and Anaphora

Writing

• Writing Output Length

Generation

• Image & Video Generation

Vision

• Vision

Agents

• Agentic Tasks