Mistral-Small-24B-Base-2501

Home/Models/mistralai/Mistral-Small-24B-Base-2501

mistralai/Mistral-Small-24B-Base-2501

kind: original
quant: bf16 (16-bit)
family: mistral
modality: —
root model: mistralai/Mistral-Small-24B-Base-2501
provider: mistralai
license: apache-2.0
language: en, fr, de, es, it, pt, zh, ja, ru, ko
context: —
library: vllm
vRAM (GB): —
size on disk: —
downloads: 3,652
likes: 264
created: —
last modified: 2025-07-28T17:24:54+00:00 (2025)
huggingface: https://huggingface.co/mistralai/Mistral-Small-24B-Base-2501
tags: vllmsafetensorsmistralmistral-commonenfrdeesitptzhjarukoarxiv:2009.03300arxiv:1905.07830arxiv:1911.11641arxiv:1904.09728arxiv:1905.10044arxiv:1907.10641arxiv:1811.00937arxiv:1809.02789arxiv:1911.01547arxiv:1705.03551

benchmark	category	score	context	conf	evidence
MMLU	knowledge	80.73	[MMLU][]	medium	`\| [MMLU][mmlu] \| 5-shot \| 80.73 \|`
GSM8K	math	80.73	[GSM8K][]	medium	`\| [GSM8K][gsm8k] \| 5-shot, maj@1 \| 80.73 \|`
TriviaQA	knowledge	80.32	[TriviaQA][]	medium	`\| [TriviaQA][triviaqa] \| 5-shot \| 80.32 \|`
MMLU	knowledge	78.86	Spanish MMLU	medium	`\| Spanish MMLU \| - \| 78.86 \|`
MMLU	knowledge	78.03	French MMLU	medium	`\| French MMLU \| - \| 78.03 \|`
MMLU	knowledge	77.69	German MMLU	medium	`\| German MMLU \| - \| 77.69 \|`
MMLU	knowledge	75.64	Russian MMLU	medium	`\| Russian MMLU \| - \| 75.64 \|`
MMLU	knowledge	74.46	Japanese MMLU	medium	`\| Japanese MMLU \| - \| 74.46 \|`
MMLU	knowledge	70.35	Chinese MMLU	medium	`\| Chinese MMLU \| - \| 70.35 \|`
MBPP	coding	69.64	[MBPP][]	medium	`\| [MBPP][mbpp] \| pass@1 \| 69.64 \|`
AGIEval	reasoning	65.8	[AGIEval][]	medium	`\| [AGIEval][agieval] \| - \| 65.80 \|`
MMLU	knowledge	56.42	Korean MMLU	medium	`\| Korean MMLU \| - \| 56.42 \|`
MMLU-Pro	knowledge	54.37	[MMLU Pro][mmlu_pro]	medium	`\| [MMLU Pro][mmlu_pro] \| 5-shot, CoT \| 54.37 \|`
MATH	math	45.98	[MATH][]	medium	`\| [MATH][math] \| 4-shot, MaJ \| 45.98 \|`
GPQA	reasoning	34.37	[GPQA Main][]	medium	`\| [GPQA Main][gpqa] \| 5-shot, CoT \| 34.37 \|`

none scraped.