Instructions to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF",
	filename="Mistral-Nemo-Instruct-12B_Q4_k_m.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Use Docker

docker model run hf.co/SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Ollama
How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with Ollama:
```
ollama run hf.co/SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
```

Unsloth Studio new

How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF to start chatting

Docker Model Runner
How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M
```

Lemonade

How to use SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SandLogicTechnologies/Mistral-NeMo-12B-Instruct-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.Mistral-NeMo-12B-Instruct-GGUF-Q4_K_M

List all available models

lemonade list

Mistral-NeMo-12B-Instruct

Mistral-NeMo-12B-Instruct is an instruction-optimized large language model built on top of the Mistral-NeMo-12B foundation. It is tailored for dialogue systems, structured task execution, and complex reasoning scenarios. The model emphasizes clarity, stability across multi-turn interactions, and high-quality output generation for both technical and general-purpose tasks.

For streamlined local deployment, GGUF quantized builds (including Q4_K_M and Q5_K_M) are commonly used to reduce memory demands and improve inference performance on consumer hardware.

Overview

Model name: Mistral-NeMo-12B-Instruct
Base architecture: Decoder-only transformer
Developer: NVIDIA & Mistral AI
License: Apache-2.0
Quantized Versions:
- Q4_K_M (4-bit)
- Q5_K_M (5-bit)
Parameter count: ~12B
Tokenizer: NeMo-aligned tokenizer
Model type: Instruction-tuned language model

Quantization Details

Q4_K_M

Approx. ~ 70% size reduction (~ 6.96)
Reduced VRAM/RAM requirements
Suitable for CPU-heavy or limited-GPU setups
Faster inference speed
Slight quality trade-off in edge cases

Q5_K_M

Approx. ~ 64% size reduction (~ 8.13)
Balanced compression and quality
Improved stability over 4-bit variants
Better retention of reasoning depth
Preferred when additional memory is available

Key Features

Instruction alignment Tuned to respond accurately to detailed prompts.
Dialogue consistency Maintains structured context across extended exchanges.
Problem-solving capability Handles analytical and multi-step reasoning tasks.
Programming support Assists with code writing, refactoring, and explanations.
Formatted output generation Capable of structured responses when required.
Extended context handling Designed to manage long prompts efficiently.

Training Details

This model originates from the Mistral-NeMo-12B base and undergoes additional refinement to enhance responsiveness and alignment with user instructions.

Pretraining: Large-scale autoregressive training on diverse multilingual and code datasets.
Architecture: 12B parameter transformer decoder model.
Instruction Tuning: Supervised fine-tuning for improved instruction following and chat performance.
Context Length: Supports up to 128K token context window.

Usage

llama.cpp

./llama-cli \
  -m SandlogicTechnologies\Mistral-NeMo-12B-Instruct_Q4_K_M.gguf \
  -p "Summarize the key differences between CNNs and Transformers."

Recommended Use Cases

Conversational systems Multi-turn assistant-style interactions.
Technical documentation support Drafting and summarizing structured material.
Code-related workflows Generating and reviewing source code.
Analytical reasoning tasks Assisting with structured logic and problem solving.
Automation pipelines Producing formatted outputs for downstream systems.

Acknowledgments

These quantized models are based on the original work by mistralai development team.

Special thanks to:

The mistralai team for developing and releasing the Mistral-NeMo-12B-Instruct model.
Georgi Gerganov and the entire llama.cpp open-source community for enabling efficient model quantization and inference via the GGUF format.

Contact

For any inquiries or support, please contact us at support@sandlogic.com or visit our Website.

Downloads last month: 94

GGUF

Model size

12B params

Architecture

llama

Hardware compatibility

4-bit

5-bit