Ministral 3 8B Instruct 2512 against Ministral 3 14B Instruct (2512)

Pick two models to see them side by side: Overall score, every category, the questions where one did better, and speed and memory. Each model is shown at its newest tested version. Back to the leaderboard.

The short version

Same model line in two sizes: Ministral 3 8B Instruct 2512 has 8 billion parameters and Ministral 3 14B Instruct (2512) has 14 billion parameters.

Ministral 3 14B Instruct (2512) scores higher overall: 92.7 against 88.7 out of 100, a gap of 4.0 points.

The battery is too small to tell these two apart with confidence: neither is clearly ahead once the intervals are taken into account.

Across the 9 categories that count toward Overall, Ministral 3 8B Instruct 2512 is higher in 0, Ministral 3 14B Instruct (2512) is higher in 5 and 4 are level.

The widest gap is in Multilingual: 88.3 for Ministral 3 8B Instruct 2512 against 98.3 for Ministral 3 14B Instruct (2512), in favour of Ministral 3 14B Instruct (2512).

Of the 259 questions both models have a definite score for, Ministral 3 8B Instruct 2512 did better on 9, Ministral 3 14B Instruct (2512) did better on 24, and 226 were level.

The two were tested in different runs, so their hard-question scores are not compared.

Their speeds were not measured on one shared backend and Ollama version, so they are not compared.

Written from the numbers below. Scores come from one fixed battery run on one laptop; read the methodology before drawing conclusions.

Overall

Ministral 3 14B Instruct (2512) scores higher: 92.7 against 88.7, a gap of 4.0 points. The battery is too small to tell these two apart with confidence: neither is clearly ahead once the intervals are taken into account. How sure is a score?

Overall (0–100) with its 95% interval, the rank range among all listed ranked models, and the hard-question score.
MeasureMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
Overall88.785.2–92.192.790.0–95.2
Rank range19–869–61
Hard questionsmean score on the questions that half or fewer of the ranked models of 3B and over in the model's own run answered correctly68.862.4
Size class4–9B9–16B

By category

Programmatic mean (0–100) per category. Higher is the model with the larger mean; a gap that rounds to 0.0 is called level. A dash means not scored, which is different from 0.
CategoryMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)HigherDifference
Reasoningcounts toward Overall 82.9 82.9 Level 0
Mathcounts toward Overall 80.0 88.0 Ministral 3 14B Instruct (2512) 8.0
Codingcounts toward Overall 78.4 87.5 Ministral 3 14B Instruct (2512) 9.1
Instruction followingcounts toward Overall 77.8 81.2 Ministral 3 14B Instruct (2512) 3.4
Knowledge & honestycounts toward Overall 96.7 96.7 Level 0
Long contextcounts toward Overall 94.4 100.0 Ministral 3 14B Instruct (2512) 5.6
Multilingualcounts toward Overall 88.3 98.3 Ministral 3 14B Instruct (2512) 10.0
Tool usecounts toward Overall 100.0 100.0 Level 0
Safetycounts toward Overall 100.0 100.0 Level 0
Pushback resistanceshown separately 100.0 100.0 Level 0
Famous gotchasshown separately 80.0 86.7 Ministral 3 14B Instruct (2512) 6.7
Writingshown separately — — — —
Self-descriptionshown separately — — — —

Speed and memory

Measured on one laptop. Decode speed is generation speed with one request at a time.
MeasureMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
Decode speed, one stream (tok/s)11.05.6
GPU memory on our server (1 to 8 answers at a time) (GB)23.830.0
Model size as loaded on our server (GB)23.830.0
BackendROCm 7.1 (Ollama 0.40.4, flash attention, 16k context; 8 parallel slots)ROCm 7.1 (Ollama 0.34.4, flash attention, 16k context; 4 parallel slots until 2026-10-07 00:45Z, 8 after)

A dash means the figure was not measured, which is different from 0. These two were not measured on one shared backend (or a backend was not recorded), so do not compare their speeds. Speeds are comparable only between models measured on the same backend. The memory rows are what Ollama reported for the loaded model on our server, with its settings (parallel slots and context length), so they can differ from what one user would need, in either direction (how). They are not the memory estimate the hardware picker uses. See also the Speed page.

Where they differ

Both models have a definite score for 259 questions. Ministral 3 8B Instruct 2512 did better on 9, Ministral 3 14B Instruct (2512) did better on 24, and 226 were level. A score gap of more than 0.001 counts as a win. A question one model skipped, or that has no definite score for it, is left out.

Questions both models have a score for, by category.
CategoryQuestionsMinistral 3 8B Instruct 2512 betterMinistral 3 14B Instruct (2512) betterLevel
Reasoning352231
Math251321
Coding303819
Instruction following302523
Knowledge & honesty301128
Long context180117
Multilingual200317
Tool use200020
Safety240024
Pushback resistance120012
Famous gotchas150114

Questions where only one got full marks

Full marks means a score of 1. The answers of public questions can be opened below. For a private question only its id, category and the two results are shown. Each result links to that question on the model's page.

Reasoning

Reasoning: 4 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
reasoning-006private fail on its page pass on its page
reasoning-025private pass on its page fail on its page
reasoning-027private fail on its page pass on its page
reasoning-032private pass on its page fail on its page

Math

Math: 4 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
math-004private fail on its page pass on its page
math-008private pass on its page fail on its page
math-019private fail on its page pass on its page
math-022private fail on its page pass on its page

Coding

Coding: 8 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
coding-009private 82% on its page pass on its page
coding-011private fail on its page pass on its page
coding-017private 30% on its page pass on its page
coding-022private pass on its page 45% on its page
coding-025private 22% on its page pass on its page
coding-026private pass on its page fail on its page
coding-029private 73% on its page pass on its page
coding-030private 27% on its page pass on its page

Instruction following

Instruction following: 6 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
instruction-009private 67% on its page pass on its page
instruction-012private pass on its page 75% on its page
instruction-013private pass on its page 50% on its page
instruction-014publicall models' answers 75% on its page pass on its page
Show both answers to instruction-014

Ministral 3 8B Instruct 2512

"Thank you for reaching out with your inquiry—I appreciate the opportunity to assist you in this matter. I will carefully review your request and provide a detailed response at my earliest convenience. Please accept my sincere regards."

Ministral 3 14B Instruct (2512)

"Thank you for your inquiry, and I appreciate the opportunity to address your request with the utmost professionalism. I have taken note of your details and will ensure a prompt and thorough response to your needs at the earliest convenience. Please find my regards"
instruction-016private 67% on its page pass on its page
instruction-019private 75% on its page pass on its page

Knowledge & honesty

Knowledge & honesty: 2 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
knowledge-019private pass on its page fail on its page
knowledge-024private fail on its page pass on its page

Long context

Long context: 1 question where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
long_context-014private fail on its page pass on its page

Multilingual

Multilingual: 2 questions where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
multilingual-005private fail on its page pass on its page
multilingual-020private 33% on its page pass on its page

Famous gotchas

Famous gotchas: 1 question where exactly one model got full marks.
QuestionMinistral 3 8B Instruct 2512Ministral 3 14B Instruct (2512)
gotchas-006private fail on its page pass on its page

The same numbers as JSON: compare.json. Answers appear there only for public questions.