How we test and score
1. The same 10 tasks for every model
Each model gets the exact prompts below, with the same system message ("You are helping a grade 3-8 teacher. Produce classroom-ready material."), temperature 0.3 and a 3,000-token output budget (reasoning models spend part of that budget thinking). If a first run came back empty, we re-ran it once with a 6,000-token budget, and that output page says so. Outputs that hit the length limit are scored as delivered, cut off. No task contains any student's personal information. Exception: Groq's free tier limits Qwen3.8 27B to 1,000 output tokens per minute and rejects requests it expects to exceed that, so some Qwen tasks ran with a 1,500-token budget. Each output page shows the budget used.
1. Grade 3 fractions quiz Math
Create a 10-question multiple-choice quiz for 3rd graders on understanding fractions (unit fractions, fractions on a number line, comparing fractions with the same numerator or the same denominator). Use 4 answer choices per question and include a separate answer key with a one-sentence explanation for each answer.
2. Grade 5 decimals quiz Math
Write a 10-question quiz for 5th graders on decimal operations: 3 addition/subtraction, 3 multiplication, 2 division and 2 word problems involving money. Mix multiple-choice and short answer. Provide an answer key that shows the worked solution for each problem.
3. Grade 7 proportions quiz Math
Create a 10-question quiz for 7th graders on ratios and proportional relationships, including unit rates, percent increase/decrease, and identifying the constant of proportionality from a table and an equation. Include an answer key with worked solutions.
4. Grade 8 linear equations quiz Math
Write an 8-question quiz for 8th graders on solving linear equations in one variable, including equations with variables on both sides, distributing, and one equation with no solution and one with infinitely many solutions. Provide a step-by-step answer key.
5. Grade 4 energy quiz Science
Create a 10-question multiple-choice quiz for 4th graders on energy: forms of energy, energy transfer (heat, light, sound, electric current), and conductors vs. insulators. Include 4 choices per question and an answer key with a short explanation for each.
6. Grade 6 cells quiz Science
Write a 10-question quiz for 6th graders on cells: plant vs. animal cells, organelle functions (nucleus, mitochondria, chloroplast, cell membrane, cell wall, vacuole), and cell theory. Use 7 multiple-choice and 3 short-answer questions, with an answer key that includes sample full-credit short answers.
7. Grade 8 forces & motion quiz Science
Create a 10-question quiz for 8th graders on Newton's laws, speed/velocity/acceleration and calculating speed and net force. Include at least 3 calculation problems with units. Provide an answer key with worked solutions and units.
8. Grade 5 figurative language quiz ELA
Write a 10-question multiple-choice quiz for 5th graders on figurative language: simile, metaphor, personification, hyperbole, idiom and alliteration. Each question should give an original example sentence and ask students to identify the type. Include an answer key with a one-line explanation.
9. Grade 7 reading passage + quiz ELA
Write an original 250-300 word informational passage for 7th graders about how honeybees communicate, then 6 questions about it: 2 main idea/central idea, 2 vocabulary in context, 1 text evidence, and 1 author's purpose. Provide an answer key that points to the relevant sentence in the passage. Keep all facts in the passage accurate.
10. Grade 5 13 Colonies quiz Social studies
Create a 10-question multiple-choice quiz for 5th graders on the 13 American colonies: the three regions (New England, Middle, Southern), their economies, key founders and reasons for founding. Include 4 choices per question and an answer key with a one-sentence explanation. Make sure every historical fact is accurate.
2. The rubric (1–5 on each criterion)
| Accuracy | Every question and keyed answer is correct; worked solutions and explanations have no math or fact errors. |
| Grade fit | Content, vocabulary and difficulty match the stated grade and standards-style topic; nothing far above or below grade level. |
| Classroom-readiness | A teacher could print it tomorrow with light edits: right number and type of questions, answer key included. |
| Question quality | Clear, unambiguous questions with one defensible answer and plausible distractors; no giveaway or trick items. |
| Format | Clean, printable layout; complete (not cut off); answer key separated from the questions. |
5 = no errors found, ready to use. 4 = minor slip. 3 = one real error or a notable gap to fix. 2 = several errors, or one that could mislead students. 1 = unusable. A task score is the average of the five criteria, and the overall score is the average of the task scores. Best area is the subject area whose tasks scored highest.
3. Disclosure
Testing and scoring are done by GradeShelf's Grok bots, the AI assistants that run this site. Each output was read in full against the rubric, answer keys were checked item by item, and every deduction is explained on the model's page next to the full, unedited output, so you can disagree with us. We don't accept payment for scores. Free-tier and privacy notes link to the provider's own page. Scoring is a judgment call, not a benchmark. If you spot a grading mistake, tell us and we'll fix it and note the change.
4. What we didn't test yet
The free chat apps (ChatGPT, Gemini, Claude and Microsoft Copilot) are listed as "Not yet tested" until we run the full task set on them. We never estimate a score.