MicroEvals

Run your prompts across multiple models to compare their performance.
History

Models

Prompt

Examples

Real-world professional tasks across occupations and sectors, by OpenAI 

Loading questions…

Public evaluations

Showing 1541-1560 of 3637

Surveygo.pro
👍0
Surveygo.pro

Melissa Cleary

Topological Rollback & Geodesic Steering Engine. Here is our...
👍0
Topological Rollback & Geodesic Steering Engine. Here is our...

test
👍0
test

Creative writing involving Confucius, Alexander the great, Elon musk and Hitler
👍0
Creative writing involving Confucius, Alexander the great, Elon musk and Hitler

Comparison
👍0
Comparison

Executive directors are responsible for running the firm. A...
👍0
Executive directors are responsible for running the firm. A...

TESTE
👍0
TESTE

Test
👍0
Test

test

Test
👍0
Test

Propolis te bulunan Fito kimyasallar ve sağlığa faydaları
👍0
Propolis te bulunan Fito kimyasallar ve sağlığa faydaları

Evsluacion de extraccion de APK
👍0
Evsluacion de extraccion de APK

Saber que modelo me puede ayudar

hello
👍0
hello

在密勒运放中，如果调零电阻Rz取值过大，会导致过补偿，表现为相位曲线翘起且UGF明显大于GBW，这常常会导致最终相位裕度...
👍0
在密勒运放中，如果调零电阻Rz取值过大，会导致过补偿，表现为相位曲线翘起且UGF明显大于GBW，这常常会导致最终相位裕度...

Efficient Task Manager in C
👍0
Efficient Task Manager in C

[7/25/2026 2:08 PM] Atamus: Create a premium educational sci...
👍0
[7/25/2026 2:08 PM] Atamus: Create a premium educational sci...

Создай премиальный, современный и максимально конверсионный ...
👍0
Создай премиальный, современный и максимально конверсионный ...

Provide me an analysis of the most recent world cup statisti...
👍0
Provide me an analysis of the most recent world cup statisti...

Knowledge
👍0
Knowledge

Asking the question to the model to act as a god ( as in fantasy novels) to provide truth about people as a visionary.

Janet’s ducks lay 16 eggs per day. She eats three for breakf...
👍0
Janet’s ducks lay 16 eggs per day. She eats three for breakf...

You are an expert AI assistant. Your task is to solve the problem below as accurately as possible. Requirements: 1. Do not make assumptions without stating them. 2. If information is insufficient, explicitly say what is missing. 3. Show concise reasoning without unnecessary verbosity. 4. Clearly separate facts from assumptions. 5. If multiple solutions exist, compare them. 6. Mention limitations or risks. 7. If you are uncertain, estimate your confidence (0–100%) and explain why. 8. Optimise for correctness over sounding confident. Problem: Imagine you have been hired as the lead technical architect for a startup developing an AI-powered medical decision support system for rare disease diagnosis. The system must: - Accept natural language symptom descriptions. - Retrieve relevant diseases from a knowledge base. - Rank candidate diseases. - Generate an evidence-backed explanation. - Minimise hallucinations. - Handle conflicting evidence. - Scale to 1 million patient queries per month. - Be explainable enough for clinicians. Your tasks: A. Design the complete architecture from user input to final response. B. Explain why each component exists and what problem it solves. C. Identify at least five potential failure modes. D. Explain how you would evaluate the system before deployment. E. Recommend the best retrieval strategy and justify your choice. F. Explain how you would reduce hallucinations. G. Discuss security, privacy, and regulatory considerations. H. Suggest improvements for a Version 2 of the system. Output format: # Executive Summary # Architecture # Component Explanations # Failure Modes # Evaluation Plan # Hallucination Mitigation # Security & Privacy # Future Improvements # Confidence Score
👍0
You are an expert AI assistant. Your task is to solve the problem below as accurately as possible. Requirements: 1. Do not make assumptions without stating them. 2. If information is insufficient, explicitly say what is missing. 3. Show concise reasoning without unnecessary verbosity. 4. Clearly separate facts from assumptions. 5. If multiple solutions exist, compare them. 6. Mention limitations or risks. 7. If you are uncertain, estimate your confidence (0–100%) and explain why. 8. Optimise for correctness over sounding confident. Problem: Imagine you have been hired as the lead technical architect for a startup developing an AI-powered medical decision support system for rare disease diagnosis. The system must: - Accept natural language symptom descriptions. - Retrieve relevant diseases from a knowledge base. - Rank candidate diseases. - Generate an evidence-backed explanation. - Minimise hallucinations. - Handle conflicting evidence. - Scale to 1 million patient queries per month. - Be explainable enough for clinicians. Your tasks: A. Design the complete architecture from user input to final response. B. Explain why each component exists and what problem it solves. C. Identify at least five potential failure modes. D. Explain how you would evaluate the system before deployment. E. Recommend the best retrieval strategy and justify your choice. F. Explain how you would reduce hallucinations. G. Discuss security, privacy, and regulatory considerations. H. Suggest improvements for a Version 2 of the system. Output format: # Executive Summary # Architecture # Component Explanations # Failure Modes # Evaluation Plan # Hallucination Mitigation # Security & Privacy # Future Improvements # Confidence Score

You are an expert AI assistant. Your task is to solve the problem below as accurately as possible. Requirements: 1. Do not make assumptions without stating them. 2. If information is insufficient, explicitly say what is missing. 3. Show concise reasoning without unnecessary verbosity. 4. Clearly separate facts from assumptions. 5. If multiple solutions exist, compare them. 6. Mention limitations or risks. 7. If you are uncertain, estimate your confidence (0–100%) and explain why. 8. Optimise for correctness over sounding confident. Problem: Imagine you have been hired as the lead technical architect for a startup developing an AI-powered medical decision support system for rare disease diagnosis. The system must: - Accept natural language symptom descriptions. - Retrieve relevant diseases from a knowledge base. - Rank candidate diseases. - Generate an evidence-backed explanation. - Minimise hallucinations. - Handle conflicting evidence. - Scale to 1 million patient queries per month. - Be explainable enough for clinicians. Your tasks: A. Design the complete architecture from user input to final response. B. Explain why each component exists and what problem it solves. C. Identify at least five potential failure modes. D. Explain how you would evaluate the system before deployment. E. Recommend the best retrieval strategy and justify your choice. F. Explain how you would reduce hallucinations. G. Discuss security, privacy, and regulatory considerations. H. Suggest improvements for a Version 2 of the system. Output format: # Executive Summary # Architecture # Component Explanations # Failure Modes # Evaluation Plan # Hallucination Mitigation # Security & Privacy # Future Improvements # Confidence Score