Rewrite · 6 checks · max 400 tokens · benched 2026-09-16

Mistral Large 3 2512 vs Mistral Medium 3.5 vs Mistral Small 4 vs Ministral 3 8B 2512 on jargon to plain english for a 12-year-old

Mistral AI models side by side on "Jargon to plain English for a 12-year-old": Ministral 3 8B 2512 scores 9/10; Ministral 3 8B 2512 is the cheapest answer scoring 8+ at $0.02 per 1,000 runs. Outputs, checks, judge reasons, latency and cost.

The prompt every model received

User

Rewrite this paragraph so a 12-year-old would understand it. Keep it under 70 words and do not use the words "leverage", "synergy" or "paradigm". Output only the rewritten paragraph.

Paragraph: Our next-generation battery architecture leverages a solid-state electrolyte paradigm to achieve synergy between energy density and thermal stability, enabling a 40% increase in range while mitigating the combustion risk profile associated with legacy lithium-ion chemistries.

Rubric for the judge: Simple vocabulary, keeps the 40% range gain and the safety point, under 70 words, banned words absent.

Side by side

Every cell is one OpenRouter call at temperature 0 with the prompt's token cap and reasoning effort "low" where the model supports it. Cost is usage × the catalogue rate in models.json. Quality is one judge call to anthropic/claude-haiku-4.5 against the prompt's rubric, cached per prompt version.

Mistral Large 3 2512

mistralai/mistral-large-2512

9/10

Latency: 519ms
Cost: $0.00012
Per 1,000: $0.12

100 in · 45 out · 35 words · checks 6/6

max_words
contains
contains
not_contains
not_contains
not_contains

Our new battery design uses special materials to make it safer and last longer. It holds more power without overheating, so electric cars can go 40% farther than old batteries—and won’t catch fire as easily.

Judge: Excellent rewrite: simple vocabulary, includes 40% range gain and safety benefit, 28 words (well under 70), avoids all banned words, and is age-appropriate for a 12-year-old.

Mistral Medium 3.5

mistralai/mistral-medium-3-5

1/10

failed: empty output (finish_reason length)

Latency: 909ms
Cost: $0.01217
Per 1,000: $12.17

112 in · 1600 out (1357 reasoning) · 0 words · checks 4/6

max_words
contains
contains
not_contains
not_contains
not_contains

(empty output)

Judge: The output is completely empty, failing to provide any rewritten paragraph despite clear task requirements.

Mistral Small 4

mistralai/mistral-small-2603

9/10

Latency: 1.1s
Cost: $0.00042
Per 1,000: $0.42

112 in · 690 out (609 reasoning) · 34 words · checks 6/6

max_words
contains
contains
not_contains
not_contains
not_contains

Our new kind of battery uses a special solid material inside it. This helps it hold more energy and stay safe from heat, letting things go 40% farther without catching fire like old batteries.

Judge: Excellent rewrite: simple vocabulary, includes 40% range gain and safety benefit, 28 words (well under 70), avoids all banned words, and is age-appropriate for a 12-year-old.

Ministral 3 8B 2512

mistralai/ministral-8b-2512

9/10

top judge scorecheapest scoring 8+

Latency: 324ms
Cost: $2.2e-5
Per 1,000: $0.02

100 in · 47 out · 38 words · checks 6/6

max_words
contains
contains
not_contains
not_contains
not_contains

Our new battery design uses a special solid middle layer to pack more power and stay cooler. This lets electric cars go 40% farther on one charge and won’t catch fire like old batteries. It’s safer and stronger!

Judge: Excellent rewrite: simple vocabulary, includes 40% range gain and safety benefits, avoids all banned words, and is 38 words (well under 70-word limit); only minor issue is 'won't catch fire' is slightly informal but still age-appropriate.

Frequently asked

What does this prompt test?

Rewrite: Simple vocabulary, keeps the 40% range gain and the safety point, under 70 words, banned words absent. The deterministic checks are max_words, contains, contains, not_contains, not_contains, not_contains.

Which model should I pick for this task?

If the judge's bar of 8/10 is good enough for you, Ministral 3 8B 2512 at $0.02 per 1,000 runs. If you need the top score, Ministral 3 8B 2512 at $0.02 per 1,000 runs.

Same prompt, other families

Run it yourself

Re-run this prompt on Mistral AI with your own OpenRouter key, or tweak the wording and see what changes.

Side by side

Mistral Large 3 2512

Mistral Medium 3.5

Mistral Small 4

Ministral 3 8B 2512

Frequently asked

Same prompt, other families

Stop overpaying for AI tools