HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks
Home/Orgs/mistralai/mistralai/Mistral-Small-3.2-24B-Instruct-2506

mistralai/Mistral-Small-3.2-24B-Instruct-2506

family mistral · 12 variants · unknown

provider
mistralai
released
—
huggingface
https://huggingface.co/mistralai/Mistral-Small-3.2-24B-Instruct-2506
best scores
WildBench: 55.6Chatbot Arena: 19.56MMLU: 80.62MMLU-Pro: 66.76MATH: 69.3GPQA: 44.42GPQA Diamond: 76.26MBPP: 74.63HumanEval: 88.99SimpleQA: 10.43MMMU: 64.0MathVista: 68.91ChartQA: 86.24DocVQA: 94.08AIME 2024: 91.82AIME 2025: 83.48LiveCodeBench: 75.0

Best scores per benchmark

benchmarkcategorybest score
DocVQAmultimodal94.08
AIME 2024math91.82
HumanEvalcoding88.99
ChartQAmultimodal86.24
AIME 2025math83.48
MMLUknowledge80.62
GPQA Diamondreasoning76.26
LiveCodeBenchcoding75.0
MBPPcoding74.63
MATHmath69.3
MathVistamultimodal68.91
MMLU-Proknowledge66.76
MMMUmultimodal64.0
WildBenchchat55.6
GPQAreasoning44.42
Chatbot Arenachat19.56
SimpleQAknowledge10.43

Variants (12)

variantkindquantbitsvRAMdlorgyear
Mistral-Small-3.2-24B-Instruct-2506-GGUFggufgguf16.039,010unsloth2025
Magistral-Small-2509originalbf1616.034,001mistralai2026
Mistral-Small-3.2-24B-Instruct-2506-MLX-6bitmlxmlx24,196lmstudio-community2025
Mistral-Small-3.2-24B-Instruct-2506-MLX-4bitmlxmlx20,450lmstudio-community2025
Mistral-Small-3.2-24B-Instruct-2506-MLX-8bitmlxmlx19,619lmstudio-community2025
mistralai_Mistral-Small-3.2-24B-Instruct-2506-GGUFggufgguf2.08,781bartowski2025
Mistral-Small-3.2-24B-Instruct-2506-GGUFggufgguf16.07,750lmstudio-community2025
Mistral-Small-3.2-24B-Instruct-2506-bnb-4bitbnb-4bit4bit6,540unsloth2025
Mistral-Small-3.2-24B-Instruct-2506originalbf1616.04,607unsloth2025
Mistral-Small-3.2-24B-Instruct-2506-unsloth-bnb-4bitbnb-4bit4bit2,730unsloth2025
Magistral-Small-2509-GGUFggufgguf16.0295Mungert2025
Mistral-Small-3.2-24B-Instruct-2506-FP8fp8fp88.074unsloth2025
Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.