Mistral-Nemo-Instruct-FP8-2407

Home/Models/mistralai/Mistral-Nemo-Instruct-FP8-2407

mistralai/Mistral-Nemo-Instruct-FP8-2407

kind: fp8
quant: fp8 (8-bit)
family: mistral
modality: —
root model: mistralai/Mistral-Nemo-Base-2407
provider: mistralai
license: apache-2.0
language: en, fr, de, es, it, pt, ru, zh, ja
context: —
library: vllm
vRAM (GB): —
size on disk: —
downloads: 115
likes: 26
created: —
last modified: 2025-07-28T17:26:56+00:00 (2025)
huggingface: https://huggingface.co/mistralai/Mistral-Nemo-Instruct-FP8-2407
tags: vllmmistral-commonenfrdeesitptruzhjabase_model:mistralai/Mistral-Nemo-Base-2407base_model:finetune:mistralai/Mistral-Nemo-Base-2407license:apache-2.0region:us

benchmark	category	score	context	conf	evidence
HellaSwag	knowledge	83.5	HellaSwag (0-shot)	medium	`\| HellaSwag (0-shot) \| 83.5% \|`
WinoGrande	knowledge	76.8	Winogrande (0-shot)	medium	`\| Winogrande (0-shot) \| 76.8% \|`
TriviaQA	knowledge	73.8	TriviaQA (5-shot)	medium	`\| TriviaQA (5-shot) \| 73.8% \|`
CommonsenseQA	knowledge	70.4	CommonSenseQA (0-shot)	medium	`\| CommonSenseQA (0-shot) \| 70.4% \|`
MMLU	knowledge	68.0	MMLU (5-shot)	medium	`\| MMLU (5-shot) \| 68.0% \|`
OpenBookQA	knowledge	60.6	OpenBookQA (0-shot)	medium	`\| OpenBookQA (0-shot) \| 60.6% \|`
TruthfulQA	safety	50.3	TruthfulQA (0-shot)	medium	`\| TruthfulQA (0-shot) \| 50.3% \|`

none scraped.