Ministral 3 14B
MistralMistralOpen WeightApache 2.0 · Commercial OK
Description
A balanced model in the Ministral 3 family, Ministral 3 14B is a powerful, efficient tiny language model with vision capabilities. This model is the instruct post-trained version in FP8, fine-tuned for instruction tasks, making it ideal for chat and instruction based use cases. The Ministral 3 family is designed for edge deployment, capable of running on a wide range of hardware. Ministral 3 14B can even be deployed locally, capable of fitting in 24GB of VRAM in BF16, and less than 12GB of RAM/VRAM when quantized.
Release Date
2025-12-02
Parameters
14.0B
Context Length
262K
Modalities
image, text
Capability Radar
25
general
23
coding
35
reasoning
35
science
31
agents
10
multimodal
Rankings
| Domain | #Rank | Score | Source |
|---|---|---|---|
| Code Ranking | 492 | 22.0 | AA |
| General Ranking | 511 | 28.0 | AA |
| Science | 472 | 30.0 | AA |
Benchmark Scores (LLM Stats)
(LLM Stats (zeroeval))Chat
MM-MT-Bench
0.08 / 100SR
General
MMLU
79.4%SR
TriviaQA
74.9%SR
Multilingual MMLU
74.2%SR
Wild Bench
68.5%SR
Arena Hard
55.1%SR
Language
MMLU-Redux
82.0%SR
Math
MATH
90.4%SR
AIME 2024
89.8%SR
AIME 2025
85.0%SR
MATH (CoT)
67.6%SR
Reasoning
GPQANYU + Cohere + Anthropic (2023)
71.2%SR
AGIEval
64.8%SR
LiveCodeBench
64.6%SR
AA Evaluation Indices
(Artificial Analysis)Mmlu Pro(TIGER-Lab (Univ. of Waterloo, Toronto, CMU, 2024))69.3
Gpqa(NYU + Cohere + Anthropic (2023))57.2
Livecodebench(UC Berkeley + MIT + Cornell (2024))35.1
Ifbench(Google Research (2023))32.0
Math Index(Artificial Analysis)30.0
Aime 25(MAA (Mathematical Association of America))30.0
Tau2(Sierra + U Toronto + Vector Institute (2025))27.2
Lcr(Artificial Analysis)26.3
Scicode(UIUC + Argonne National Lab (2024))23.8
Coding Index(Artificial Analysis)14.4
Terminalbench V2 19.7
Tau Banking6.6
Intelligence Index(Artificial Analysis)6.0
Hle(Center for AI Safety + Scale AI (2025))4.6
Terminalbench Hard(Stanford × Laude Institute (2026))4.5
LLM Stats Category Scores
(LLM Stats (zeroeval))Math90
Reasoning70
General60
Creativity60
Writing60
Communication40
Chat30
Multimodal10
Pricing
Input Price$0.2 / 1M tokens
Output Price$0.2 / 1M tokens
Blended Price (3:1)$0.2 / 1M tokens
Cache Read Price$0.02 / 1M tokens
Speed
Tokens/sec85.3
Time to First Token0.55s
Time to Answer0.55s
Provider Price Ranking
Provider Price Ranking
4 providers
Cheapest: MistralMost Expensive: Kilo Gateway
ProviderInputOutput
1MistralPRIMARY
$0.2
$0.2
2NanoGPT
$0.2
$0.2
3OpenRouter
$0.2
$0.2
4Kilo Gateway
$0.2
$0.2
Compare pricing across different API providers for this model.