Ministral 3B (Dec '25)

Benchmark scores, pricing, speed and comparisons for Ministral 3B (Dec '25).

Ministral 3B (Dec '25) is an AI model from Mistral, released 2025-12-02. It scores 12.1 on the Intelligence Index, 4.8 on the Coding Index and 22.0 on the Math Index.

Ministral 3B (Dec '25) at a glance

Mistral released Ministral 3B (Dec '25) on December 2, 2025. It is marked as retired in the catalogue, so treat its numbers as a historical record.

Ministral 3B (Dec '25) scores 12.1 on the Intelligence Index, the overall score the AI Benchmark rankings sort models by. On the narrower indexes it scores 4.8 for coding and 22.0 for math. Of its 10 individual benchmark results, the highest is 52.4% on MMLU Pro and the lowest is 0.0% on TerminalBench Hard. These scores come from independent third-party model evaluations, which Franklin AI syncs automatically each day.

It costs $0.10 per million input tokens and $0.10 per million output tokens, so a sample workload of 1 million input tokens and 200,000 output tokens costs about $0.12. It writes about 292 output tokens per second, so a 500-token answer takes roughly 1.7 seconds once output starts, after a wait of about 0.28 seconds for the first token.

Every pair of models in the catalogue has a head-to-head page. Pick Ministral 3B (Dec '25) and a rival on the model comparisons page to compare their scores, prices and speed side by side.

Key scores

Intelligence12.1Index score
Coding4.8Index score
Math22.0Index score
Speed292.4Tokens/sec
Blended Cost$0.10Per 1M tokens

How does Ministral 3B (Dec '25) score on each benchmark?

10 benchmark results for Ministral 3B (Dec '25).

MMLU Pro Knowledge
52.4%
GPQA Reasoning
35.8%
IFBench Other
26.8%
TAU2 Other
24.9%
LiveCodeBench Code
24.7%
AIME 2025 Math
22.0%
SciCode Other
14.4%
LCR Other
11.7%
HLE Other
5.3%
TerminalBench Hard Other
0.0%