Is Ministral 3 14B 2512 better than Mistral Small 4?

On accuracy they are statistically tied: Ministral 3 14B 2512 scored 33.8% and Mistral Small 4 32.5% on the same 80 questions from open datasets. Similar accuracy; Ministral 3 14B 2512 costs less per question.

Which is cheaper, Ministral 3 14B 2512 or Mistral Small 4?

Ministral 3 14B 2512 is cheaper: $0.00031 per question as billed, against $0.0019. List prices: Ministral 3 14B 2512 $0.2 in / $0.2 out, Mistral Small 4 $0.15 in / $0.6 out per million tokens.

Which is faster, Ministral 3 14B 2512 or Mistral Small 4?

Mistral Small 4 is faster at 170 output tokens per second, against 32. Time to first token: 0.6 s for Ministral 3 14B 2512, 0.5 s for Mistral Small 4.

Which has the larger context window, Ministral 3 14B 2512 or Mistral Small 4?

Both accept about the same amount of context: 262k and 262k tokens.

Head to head · tested 27 Oct 2026

Ministral 3 14B 2512 vs Mistral Small 4: accuracy, speed and price

Similar accuracy; Ministral 3 14B 2512 costs less per question.

MistralMinistral 3 14B 251233.8% score · #61$0.00031 per question32 tok/s

MistralMistral Small 432.5% score · #62$0.0019 per question170 tok/s

Verdict

Built from the measurements below

Accuracy is too close to call: Ministral 3 14B 2512 scored 33.8% and Mistral Small 4 32.5% on the same 80 questions, a gap within what chance alone could produce.
Ministral 3 14B 2512 is 6.0× cheaper per question as billed ($0.00031 vs $0.0019), reasoning tokens included.
Mistral Small 4 generates text 5.3× faster: 170 tokens per second against 32.
Mistral Small 4 starts answering sooner: 0.5 s to the first token, against 0.6 s for Ministral 3 14B 2512.

Which should you choose?

Choose Ministral 3 14B 2512 for the lower cost per correct answer ($0.00092 against $0.0057) and shorter outputs (about 1,255 tokens per question against 3,021). Choose Mistral Small 4 for faster streaming and a quicker first token. The widest subject gap is in economics: Mistral Small 4 answered 6 of 9 correctly, Ministral 3 14B 2512 4 of 9. With so few questions per subject, treat this as a lead to test on your own prompts, not a result.

Accuracy gap +1.3 points (z = 0.17). A winner is called only when z is at least 1.96, the usual 95% threshold. Speed, latency and price need a 10% margin.

Head to head

Same questions, same prompt, same API

Measure	Ministral 3 14B 2512	Mistral Small 4	Edge
Verification scoreHigher is better	33.8%	32.5%	Too close to call
Cost per questionLower is better	$0.00031	$0.0019	Ministral 3 14B 2512
Output speedHigher is better	32	170	Mistral Small 4
Time to first tokenLower is better	0.6 s	0.5 s	Mistral Small 4
End-to-end response timeLower is better	12.7 s	4.5 s	Mistral Small 4
Tokens per questionLower is better	1,255	3,021	Ministral 3 14B 2512
Context windowHigher is better	262k	262k	Similar

Side by side

Bars are scaled to the larger of the two values. The winner on each measure is marked in bold.

Every measure

Ministral 3 14B 2512 first, Mistral Small 4 second

Verification scorehigher is better

Ministral 3 14B 251233.8%

Mistral Small 432.5%

Cost per questionlower is better

Ministral 3 14B 2512$0.00031

Mistral Small 4$0.0019

Output speedhigher is better

Ministral 3 14B 251232

Mistral Small 4170

Time to first tokenlower is better

Ministral 3 14B 25120.6 s

Mistral Small 40.5 s

End-to-end response timelower is better

Ministral 3 14B 251212.7 s

Mistral Small 44.5 s

Tokens per questionlower is better

Ministral 3 14B 25121,255

Mistral Small 43,021

Context windowhigher is better

Ministral 3 14B 2512262k

Mistral Small 4262k

Score vs. cost per question

Both models in color, every other model in grey

Results by subject and dataset

Questions answered correctly. Small samples: a gap of one or two questions is noise.

Subject	Questions	Ministral 3 14B 2512	Mistral Small 4
Science	30	8/30	7/30
Engineering	15	2/15	0/15
Medicine	11	6/11	5/11
Economics	9	4/9	6/9
Law	6	2/6	2/6
Humanities & social science	9	5/9	6/9
SuperGPQA dataset	50	8/50	7/50
MMLU-Pro dataset	30	19/30	19/30

Specifications and pricing

From OpenRouter's catalogue, plus the leaderboard's test dates

Specification	Ministral 3 14B 2512	Mistral Small 4
Creator	Mistral	Mistral
Released	2 Dec 2025	16 Mar 2026
Weights	Open weights	Open weights
Reasoning	No	Yes
Input	text, image	text, image
Output	text	text
Context window	262k tokens	262k tokens
Knowledge cutoff	—	—
List price in / out per 1M	$0.2 / $0.2	$0.15 / $0.6
Blended price per 1M (3:1)	$0.20	$0.26
Cheapest host	Mistral · $0.2 / $0.2	Mistral · $0.15 / $0.6
Hosts on OpenRouter	1	1
Last tested	26 Oct 2026	26 Oct 2026

Compare any two models →

Ministral 3 14B 2512 vs Mistral Small 4: frequently asked questions

Is Ministral 3 14B 2512 better than Mistral Small 4?: On accuracy they are statistically tied: Ministral 3 14B 2512 scored 33.8% and Mistral Small 4 32.5% on the same 80 questions from open datasets. Similar accuracy; Ministral 3 14B 2512 costs less per question.
Which is cheaper, Ministral 3 14B 2512 or Mistral Small 4?: Ministral 3 14B 2512 is cheaper: $0.00031 per question as billed, against $0.0019. List prices: Ministral 3 14B 2512 $0.2 in / $0.2 out, Mistral Small 4 $0.15 in / $0.6 out per million tokens.
Which is faster, Ministral 3 14B 2512 or Mistral Small 4?: Mistral Small 4 is faster at 170 output tokens per second, against 32. Time to first token: 0.6 s for Ministral 3 14B 2512, 0.5 s for Mistral Small 4.
Which has the larger context window, Ministral 3 14B 2512 or Mistral Small 4?: Both accept about the same amount of context: 262k and 262k tokens.