IntermediateAnalysis~10 min

Model vs Model

Compare two AI models across five dimensions — using a structured format that would help a product team choose between them.

Skills Tested

Model comparisonStructured evaluationProduct decision support

Challenge Prompt

This is the example prompt structure. Write your own variation below.

Compare GPT-4o and Claude 3.5 Sonnet for a B2B customer support automation use case. Dimensions: (1) response latency, (2) context window and recall, (3) instruction following reliability, (4) cost per 1M tokens, (5) safety and refusal behaviour. For each dimension: score each model 1-5, write one sentence of evidence, and declare a winner. Final row: overall recommendation with one-sentence justification. Output as a comparison table.

Your Prompt

Write your own version. Every word matters — be specific.

🪙

Reward

+100 GeraCoins

Earn GeraCoins for completing this challenge. Redeem across all Gera products.

Hints (3)

1Use public benchmark data where available — ask the AI to cite sources.
2"One sentence of evidence" keeps the table scannable while grounding scores.
3The recommendation row is the most important — it should synthesise the whole table.