Instructions to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF",
	filename="Mellum2-12B-A2.5B-Instruct_IQ3_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-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/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_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/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
# Run inference directly in the terminal:
./llama-cli -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_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/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Use Docker

docker model run hf.co/SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

LM Studio
Jan

vLLM

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SandLogicTechnologies/Mellum2-12B-A2.5B-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/Mellum2-12B-A2.5B-Instruct-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Ollama
How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with Ollama:
```
ollama run hf.co/SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
```

Unsloth Studio

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-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/Mellum2-12B-A2.5B-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/Mellum2-12B-A2.5B-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/Mellum2-12B-A2.5B-Instruct-GGUF to start chatting

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with Pi:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "llama-cpp": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with Hermes Agent:

Start the llama.cpp server

# Install llama.cpp:
brew install llama.cpp
# Start a local OpenAI-compatible server:
llama-server -hf SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Run Hermes

hermes

Atomic Chat new
Docker Model Runner
How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M
```

Lemonade

How to use SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF:IQ3_M

Run and chat with the model

lemonade run user.Mellum2-12B-A2.5B-Instruct-GGUF-IQ3_M

List all available models

lemonade list

Mellum2-12B-A2.5B-Instruct

Mellum2-12B-A2.5B-Instruct is an instruction-tuned language model developed by JetBrains and optimized for software engineering workflows, code generation, code understanding, technical assistance, and structured text generation. This repository contains GGUF quantized variants of the model optimized for efficient local inference using llama.cpp.

Built upon the Mellum2 architecture, the model is designed to assist developers with programming-related tasks while maintaining strong instruction-following capabilities. The quantized formats significantly reduce memory requirements while preserving coding performance and technical reasoning quality, enabling practical deployment across consumer hardware and local development environments.

Model Overview

Model Name: Mellum2-12B-A2.5B-Instruct
Base Model: JetBrains/Mellum2-12B-A2.5B-Instruct
Architecture: Mixture-of-Experts (MoE) Transformer
Parameter Count: 12 Billion Total Parameters / 2.5 Billion Active Parameters
Experts: 64 Experts / 8 Active Experts per Token
Context Window: 131K Tokens
Modalities: Text
Primary Languages: English
Developer: JetBrains
License: Apache 2.0

Quantization Formats

This repository provides various GGUF quantized versions of the Mellum2-12B-A2.5B-Instruct model, optimized for efficient local inference using llama.cpp. Below are the details of the available I-Matrix (IQ) formats.

IQ3_M

Size reduction of approx 75.71% (5.50 GB) compared to 16-bit (22.64 GB)
Aggressive 3-bit quantization optimized for maximum memory efficiency
Suitable for CPU-only inference and low-memory deployment environments
Enables practical execution of coding and instruction-following workloads on constrained hardware
Output quality may reduce on complex code generation, repository-level understanding, and long-context development tasks

IQ4_NL

Size reduction of approx 71.25% (6.51 GB) compared to 16-bit (22.64 GB)
Advanced 4-bit non-linear quantization designed to better preserve coding quality and instruction-following performance
More suitable for software engineering workflows, code generation, debugging assistance, and technical tasks
Designed to reduce quantization loss compared to more aggressive formats
Slightly increased computational overhead during inference

IQ4_XS

Size reduction of approx 72.31% (6.27 GB) compared to 16-bit (22.64 GB)
Balanced 4-bit quantization focused on efficiency and stable coding performance
Good trade-off between model size, generation quality, and inference speed
Suitable for code completion, code explanation, technical assistance, and general-purpose developer workflows
Maintains reliable generation behavior across most practical software engineering workloads

Training Background (Original Model)

Mellum2-12B-A2.5B-Instruct is trained with an emphasis on software engineering, instruction following, code understanding, and practical developer-assistance workflows.

Pretraining

Large-scale training across programming languages, source code repositories, and technical documentation
Focus on code understanding, code generation, and software engineering tasks
Optimized for downstream coding and developer-assistance workloads

Instruction Tuning

Refined using instruction-following and developer-oriented datasets
Enhanced for structured responses and technical assistance workflows
Improved consistency for code generation, debugging, explanation, and software development tasks

Key Capabilities

Instruction Following Handles developer prompts and produces structured, context-aware responses.
Code Generation Supports generation of code across multiple programming languages and development workflows.
Code Understanding Assists with explaining, reviewing, and understanding existing codebases.
Developer Assistance Supports debugging, refactoring, implementation guidance, and software engineering workflows.
Efficient Local Deployment Quantized variants enable practical offline inference on consumer hardware.
Technical Text Generation Suitable for documentation, code explanation, technical Q&A, and structured outputs.

Usage Example

Using llama.cpp

./llama-cli \
  -m SandlogicTechnologies/Mellum2-12B-A2.5B-Instruct_IQ4_NL.gguf \
  -p "Create a Python function that validates JSON input and explain the implementation."

Recommended Usecases

Code Generation Workflows Generate functions, classes, scripts, and application components across multiple programming languages.
Developer Productivity Tools Support IDE integrations, local coding assistants, and software engineering workflows.
Code Review and Analysis Assist with understanding, reviewing, and improving existing code.
Technical Documentation Generate explanations, documentation, and developer-oriented content.
Research and Experimentation Evaluate coding models, prompting strategies, and local inference workflows.

Acknowledgments

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

Special thanks to:

The JetBrains team for developing and releasing the Mellum2-12B-A2.5B-Instruct model.
Georgi Gerganov and the llama.cpp open-source community for enabling efficient quantization and inference via the GGUF format.

Contact

For questions, feedback, or support, please reach out at support@sandlogic.com or visit https://www.sandlogic.com/

Downloads last month: 14

GGUF

Model size

12B params

Architecture

mellum

Hardware compatibility

3-bit

4-bit

Model tree for SandLogicTechnologies/Mellum2-12B-A2.5B-Instruct-GGUF

Base model

JetBrains/Mellum2-12B-A2.5B-Instruct

Quantized

(15)

this model