Instructions to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF",
	filename="DeepSeek-Coder-V2-Lite-Instruct_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/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf SandLogicTechnologies/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf SandLogicTechnologies/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M

Use Docker

docker model run hf.co/SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "SandLogicTechnologies/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M

Ollama
How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with Ollama:
```
ollama run hf.co/SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
```

Unsloth Studio new

How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-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/DeepSeek-Coder-V2-Lite-Instruct-GGUF to start chatting

Docker Model Runner
How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with Docker Model Runner:
```
docker model run hf.co/SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M
```

Lemonade

How to use SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.DeepSeek-Coder-V2-Lite-Instruct-GGUF-Q4_K_M

List all available models

lemonade list

Quantized DeepSeek-Coder-V2-Lite-Instruct Models

This repository provides optimized GGUF quantized versions of the DeepSeek-Coder-V2-Lite-Instruct model. These lightweight 4-bit and 5-bit variants maintain the strong reasoning and code generation capabilities of the original model, while drastically reducing compute and memory requirements, ideal for local inference or edge deployment.

Model Overview

Original Model: DeepSeek-Coder-V2-Lite-Instruct
Quantized Versions:
- Q4_K_M (4-bit quantization)
- Q5_K_M (5-bit quantization)
Architecture: Decoder-only Transformer
Base Model: DeepSeek-Coder-V2-Lite
Modalities: Text only
Developer: DeepSeek-AI
License: deepseek-license
Language: English, Chinese

Quantization Details

Q4_K_M Version

Approx. 9.66 GB memory footprint around 70% model size reduction
Best suited for consumer GPUs or CPU-based inference
Slight accuracy trade-off for maximum efficiency

Q5_K_M Version

Approx. 11.04 GB footprint around 65% size reduction
Near full precision quality retention
Ideal for scenarios requiring balanced speed and accuracy

Key Features

State-of-the-art reasoning and code generation performance
Optimized for multi-turn code completion and debugging assistance
Fine-tuned on extensive programming and instruction datasets
Strong performance in reasoning, debugging, and natural language coding tasks
Supports long-context generation up to 32K tokens

Usage

This model is suited for developers and researchers working on program synthesis, code explanation, or interactive coding assistants.

llama.cpp (text-only)

./llama-cli -hf SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF -p "Write a Python function to reverse a linked list."

Model Data

Dataset Overview

The DeepSeek-Coder-V2-Lite-Instruct model was trained on large-scale, high-quality open datasets consisting of:
- General programming languages (Python, C++, Java, Rust, etc.)
- Instruction-following data for reasoning and conversation
- StackExchange and competitive coding datasets
- Synthetic reasoning datasets for enhanced contextual understanding

Recommended Use Cases

AI Coding Assistants : - Build real-time code assistants with low-latency responses.
Code Analysis & Debugging: - Detect, explain, or correct programming errors.
Educational Platforms : - Support interactive programming tutorials and practice systems.
Edge & Low-resource Deployment : - Run code-capable LLMs on devices with limited memory (e.g., laptops, Jetson, Raspberry Pi).

Acknowledgments

These quantized models are derived from the original DeepSeek-Coder-V2-Lite-Instruct developed by DeepSeek-AI.

Special thanks to:

The deepseek team for developing and releasing the DeepSeek-Coder-V2-Lite-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: 67

GGUF

Model size

16B params

Architecture

deepseek2

Hardware compatibility

4-bit

5-bit

Model tree for SandLogicTechnologies/DeepSeek-Coder-V2-Lite-Instruct-GGUF

Base model

deepseek-ai/DeepSeek-Coder-V2-Lite-Instruct

Quantized

(64)

this model