Ministral-8B-Instruct-2410

Home/Models/mistralai/Ministral-8B-Instruct-2410

mistralai/Ministral-8B-Instruct-2410

kind: original
quant: bf16 (16-bit)
family: mistral
modality: —
root model: mistralai/Ministral-8B-Instruct-2410
provider: mistralai
license: other
language: en, fr, de, es, it, pt, zh, ja, ru, ko
context: 128k
library: vllm
vRAM (GB): —
size on disk: —
downloads: 324,344
likes: 584
created: —
last modified: 2025-07-31T10:39:07+00:00 (2025)
huggingface: https://huggingface.co/mistralai/Ministral-8B-Instruct-2410
tags: vllmsafetensorsmistralmistral-commonenfrdeesitptzhjarukolicense:othereval-resultsregion:us

benchmark	category	score	conf	evidence
WinoGrande	knowledge	74.2	low	`\| Mistral 7B Base \| 62.5 \| 42.5 \| 74.2 \| 67.9 \| 62.5 \|`
MMLU	knowledge	62.5	low	`\| Mistral 7B Base \| 62.5 \| 42.5 \| 74.2 \| 67.9 \| 62.5 \|`
TriviaQA	knowledge	62.5	low	`\| Mistral 7B Base \| 62.5 \| 42.5 \| 74.2 \| 67.9 \| 62.5 \|`
MBPP	coding	50.2	low	`\| Mistral 7B Instruct v0.3 \| 50.2 \| 38.4 \| 13.2 \|`
Chatbot Arena	chat	44.3	low	`\| Mistral 7B Instruct v0.3 \| 6.7 \| 44.3 \| 33.1 \|`
AGIEval	reasoning	42.5	low	`\| Mistral 7B Base \| 62.5 \| 42.5 \| 74.2 \| 67.9 \| 62.5 \|`
WildBench	chat	33.1	low	`\| Mistral 7B Instruct v0.3 \| 6.7 \| 44.3 \| 33.1 \|`
GSM8K	math	32.0	low	`\| Mistral 7B Base \| 26.8 \| 32.0 \|`
HumanEval	coding	26.8	low	`\| Mistral 7B Base \| 26.8 \| 32.0 \|`
MATH	math	13.2	low	`\| Mistral 7B Instruct v0.3 \| 50.2 \| 38.4 \| 13.2 \|`
MT-Bench	chat	6.7	low	`\| Mistral 7B Instruct v0.3 \| 6.7 \| 44.3 \| 33.1 \|`

none scraped.