HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Ministral-3-14B-Instruct-2512

unsloth/Ministral-3-14B-Instruct-2512

kind
original
quant
bf16 (16-bit)
family
mistral
modality
—
root model
mistralai/Ministral-3-14B-Instruct-2512
provider
unsloth
license
apache-2.0
language
en, fr, es, de, it, pt, nl, zh, ja, ko, ar
context
—
library
vllm
vRAM (GB)
—
size on disk
—
downloads
1,303
likes
7
created
—
last modified
2025-12-06T08:17:20+00:00 (2025)
huggingface
https://huggingface.co/unsloth/Ministral-3-14B-Instruct-2512
tags
vllmsafetensorsmistral3mistral-commonmistralunslothenfresdeitptnlzhjakoarbase_model:mistralai/Ministral-3-14B-Instruct-2512base_model:finetune:mistralai/Ministral-3-14B-Instruct-2512license:apache-2.0region:us

Benchmark scores (12)

benchmarkcategoryscorecontextconfevidence
MATHmath87.0low| Qwen3 14B (Non-Thinking) | 0.427 | 65.1 | 0.870 | NOT MULTIMODAL |
AIME 2024math83.7low| Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
TriviaQAknowledge74.9MATH CoT 2-Shotlow| **Ministral 3 14B** | 0.742 | <u>0.676</u> | 0.648 | 0.820 | 0.794 | 0.749 |
MMLUknowledge74.2MATH CoT 2-Shotlow| **Ministral 3 14B** | 0.742 | <u>0.676</u> | 0.648 | 0.820 | 0.794 | 0.749 |
AIME 2025math73.7low| Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
GPQA Diamondreasoning66.3low| Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
WildBenchchat65.1low| Qwen3 14B (Non-Thinking) | 0.427 | 65.1 | 0.870 | NOT MULTIMODAL |
AGIEvalreasoning64.8MATH CoT 2-Shotlow| **Ministral 3 14B** | 0.742 | <u>0.676</u> | 0.648 | 0.820 | 0.794 | 0.749 |
MATHmath62.0MATH CoT 2-Shotlow| Qwen3 14B Base | <u>0.754</u> | 0.620 | <u>0.661</u> | <u>0.837</u> | <u>0.804</u> | 0.703 |
LiveCodeBenchcoding59.3low| Qwen3-14B (Thinking) | 0.737 | 0.837 | 0.663 | 0.593 |
Chatbot Arenachat42.7low| Qwen3 14B (Non-Thinking) | 0.427 | 65.1 | 0.870 | NOT MULTIMODAL |
MT-Benchchat6.7low| Gemma3-12B-Instruct | 0.436 | 63.2 | 0.854 | 6.70 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.