What are the hardware requirements for Mistral Nemo Instruct 2407?

Mistral Nemo Instruct 2407 requires approximately 24.5 GB of VRAM and supports a context window of 128K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Mistral Nemo Instruct 2407 and how large is it?

Mistral Nemo Instruct 2407 is developed by mistralai, a model with 12.2b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Mistral Nemo Instruct 2407?

Mistral Nemo Instruct 2407 is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Mistral Nemo Instruct 2407 by mistralai — VRAM 24.5GB, 128K context

Name: Mistral Nemo Instruct 2407
Author: mistralai

Mistral Nemo Instruct 2407 is an open-source language model by mistralai. Features: 12.2b LLM, VRAM: 24.5GB, Context: 128K, License: MNPL-0.1, Instruction-Based, LLM Explorer Score: 0.33.

Base model:finetune:mistralai/... Base model:mistralai/mistral-n... De Deploy:azure En Es Eval-results Fr Instruct It Ja Mistral Mistral-common Pt Region:us Ru Safetensors Sharded Tensorflow Vllm Zh

Model Card on HF 🤗: https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407

Mistral Nemo Instruct 2407 Parameters and Internals

Model Type

instruct fine-tuned

Use Cases

Areas:

research, multilingual tasks, code tasks

Limitations:

No moderation mechanisms

Considerations:

Model should finely respect guardrails for environments requiring moderated outputs.

Additional Notes

Trained jointly by Mistral AI and NVIDIA. Drop-in replacement of Mistral 7B.

Supported Languages

en (English), fr (French), de (German), es (Spanish), it (Italian), pt (Portuguese), ru (Russian), zh (Chinese), ja (Japanese)

Training Details

Methodology:

Trained on a large proportion of multilingual and code data

Context Length:

128000

Model Architecture:

Transformer model with 40 layers, Dim: 5,120, Head dim: 128, Hidden dim: 14,336, Activation Function: SwiGLU, Number of heads: 32, Number of kv-heads: 8 (GQA), Vocabulary size: 128k (2^17), Rotary embeddings (theta = 1M)

LLM Name	Mistral Nemo Instruct 2407
Repository 🤗	https://huggingface.co/mistralai/Mistral-Nemo-Instruct-2407
Base Model(s)	Mistral Nemo Base 2407 mistralai/Mistral-Nemo-Base-2407
Model Size	12.2b
Required VRAM	24.5 GB
Updated	2026-07-13
Maintainer	mistralai
Instruction-Based	Yes
Model Files	24.5 GB 4.9 GB: 1-of-5 4.9 GB: 2-of-5 4.9 GB: 3-of-5 4.9 GB: 4-of-5 4.9 GB: 5-of-5
Supported Languages	en fr de es it pt ru zh ja
Model Architecture	MambaSSM
License	MNPL-0.1
Context Length	131072
Model Max Length	131072
Tokenizer Class	PreTrainedTokenizerFast
Vocabulary Size	131072

Rank the Mistral Nemo Instruct 2407 capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Quantized Models of the Mistral Nemo Instruct 2407

Model	Likes	Downloads	VRAM
Geneva 12B GCv2 5M	13	29	24 GB
NemoR	0	24	24 GB
...omix Unleashed 12B V0.6.1 8bit	2	4	12 GB
DeepOpus 1 12B Preview	0	18	24 GB
DeepNeo 1 12B Preview	0	15	24 GB
Ava 1.0 12B	4	34	24 GB
...istral Nemo Instruct 2407 GGUF	66	1139	4 GB
...Horizon AI Korean Advanced 12B	1	127	24 GB
...orean Mistral Nemo Sft DPO 12B	8	288	24 GB

Best Alternatives to Mistral Nemo Instruct 2407

Best Alternatives	Context / RAM	Downloads	Likes
... Instruct 2407 Absolute Heresy	128K / 24.5 GB	119	4
MistralAI Nemo Instruct ChatML	0K / 24.5 GB	3	2
...Nemo Instruct 2407 Abliterated	0K / 24.5 GB	86	3