Instructions to use DhruvalLabs/stable-code-3b-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use DhruvalLabs/stable-code-3b-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="DhruvalLabs/stable-code-3b-GGUF")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("DhruvalLabs/stable-code-3b-GGUF", dtype="auto")

llama-cpp-python

How to use DhruvalLabs/stable-code-3b-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="DhruvalLabs/stable-code-3b-GGUF",
	filename="stable-code-3b-Q2_K.gguf",
)

output = llm(
	"Once upon a time,",
	max_tokens=512,
	echo=True
)
print(output)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use DhruvalLabs/stable-code-3b-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DhruvalLabs/stable-code-3b-GGUF:Q4_K_M

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama cli -hf DhruvalLabs/stable-code-3b-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 DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf DhruvalLabs/stable-code-3b-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 DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf DhruvalLabs/stable-code-3b-GGUF:Q4_K_M

Use Docker

docker model run hf.co/DhruvalLabs/stable-code-3b-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use DhruvalLabs/stable-code-3b-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "DhruvalLabs/stable-code-3b-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DhruvalLabs/stable-code-3b-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/DhruvalLabs/stable-code-3b-GGUF:Q4_K_M

SGLang

How to use DhruvalLabs/stable-code-3b-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "DhruvalLabs/stable-code-3b-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DhruvalLabs/stable-code-3b-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "DhruvalLabs/stable-code-3b-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "DhruvalLabs/stable-code-3b-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use DhruvalLabs/stable-code-3b-GGUF with Ollama:
```
ollama run hf.co/DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
```

Unsloth Studio

How to use DhruvalLabs/stable-code-3b-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 DhruvalLabs/stable-code-3b-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 DhruvalLabs/stable-code-3b-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for DhruvalLabs/stable-code-3b-GGUF to start chatting

Atomic Chat new
Docker Model Runner
How to use DhruvalLabs/stable-code-3b-GGUF with Docker Model Runner:
```
docker model run hf.co/DhruvalLabs/stable-code-3b-GGUF:Q4_K_M
```

Lemonade

How to use DhruvalLabs/stable-code-3b-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull DhruvalLabs/stable-code-3b-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.stable-code-3b-GGUF-Q4_K_M

List all available models

lemonade list

stable-code-3b-GGUF / README.md

Dhruval

Upload LLM GGUF quants via quant-kit (batch 1)

9f61448 verified 23 days ago

preview code

Raw

History Blame Contribute Delete

5.99 kB

	---
	license: other
	base_model: stabilityai/stable-code-3b
	pipeline_tag: text-generation
	tags:
	- dataset:bigcode/the-stack-github-issues
	- dataset:meta-math/MetaMathQA
	- transformers
	- code
	- arxiv:2307.09288
	- arxiv:2309.12284
	- safetensors
	- arxiv:1910.02054
	- arxiv:2310.10631
	- stablelm
	- quantized
	- arxiv:2305.06161
	- en
	- model-index
	- dataset:bigcode/commitpackft
	- gguf
	- text-generation
	- arxiv:2204.06745
	- dataset:bigcode/starcoderdata
	- region:us
	- dataset:EleutherAI/proof-pile-2
	- dataset:tiiuae/falcon-refinedweb
	- arxiv:2104.09864
	- causal-lm
	language:
	- en
	---

	<div align="center">

	# stable-code-3b — GGUF Quantizations

	[![Model on HF](https://img.shields.io/badge/🤗-Model_on_HuggingFace-yellow)](https://huggingface.co/Dhptl/stable-code-3b-GGUF)
	[![Original Model](https://img.shields.io/badge/Original-stabilityai_stable-code-3b-blue)](https://huggingface.co/stabilityai/stable-code-3b)
	[![quant-kit](https://img.shields.io/badge/Made_with-quant--kit-green)](https://github.com/DhruvalPtl/quant-kit)

	Quantized GGUF versions of [stabilityai/stable-code-3b](https://huggingface.co/stabilityai/stable-code-3b)

	Works with [llama.cpp](https://github.com/ggerganov/llama.cpp) · [Ollama](https://ollama.ai) · [LM Studio](https://lmstudio.ai) · [Open WebUI](https://openwebui.com) · [Jan](https://jan.ai)

	Quantized by [Dhptl](https://huggingface.co/Dhptl)* on June 18, 2026 using [quant-kit](https://github.com/DhruvalPtl/quant-kit)*

	</div>

	---

	## ⚖️ The Pareto Frontier — Efficiency vs Intelligence

	> Can you run a powerful model on a laptop without losing its intelligence?

	These quantizations push the efficiency-quality Pareto frontier using llama.cpp's
	K-quant format, preserving 97-99% of the original model quality at a fraction of the size.

	\| Benchmark \| Original (FP16) \| Q4_K_M \| Quality Retained \|
	\|---\|---\|---\|---\|
	\| MMLU Pro \| See [original card](https://huggingface.co/stabilityai/stable-code-3b) \| Run benchmarks \| ~97-99% \|
	\| HellaSwag \| See [original card](https://huggingface.co/stabilityai/stable-code-3b) \| Run benchmarks \| ~97-99% \|
	\| ARC Challenge \| See [original card](https://huggingface.co/stabilityai/stable-code-3b) \| Run benchmarks \| ~97-99% \|
	\| TruthfulQA \| See [original card](https://huggingface.co/stabilityai/stable-code-3b) \| Run benchmarks \| ~97-99% \|
	\| GSM8K \| See [original card](https://huggingface.co/stabilityai/stable-code-3b) \| Run benchmarks \| ~97-99% \|


	---

	## 📦 Available Files

	\| Filename \| Size \| RAM Required \| Quant \| Quality \| Best For \|
	\|---\|---\|---\|---\|---\|---\|
	\| `stable-code-3b-Q2_K.gguf` \| 1.01 GB \| ~2.5 GB \| `Q2_K` \| ⭐ \| Extreme compression, significant quality loss. \|
	\| `stable-code-3b-Q3_K_L.gguf` \| 1.40 GB \| ~2.9 GB \| `Q3_K_L` \| ⭐⭐⭐ \| Slightly better than Q3_K_M, still a compromise. \|
	\| `stable-code-3b-Q3_K_M.gguf` \| 1.30 GB \| ~2.8 GB \| `Q3_K_M` \| ⭐⭐⭐ \| Very small file. Quality drop noticeable. \|
	\| `stable-code-3b-Q3_K_S.gguf` \| 1.17 GB \| ~2.7 GB \| `Q3_K_S` \| ⭐⭐ \| Very high compression, high quality loss. \|
	\| `stable-code-3b-Q4_K_M.gguf` \| 1.59 GB \| ~3.1 GB \| `Q4_K_M` ✅ Recommended \| ⭐⭐⭐⭐ \| Best balance of size and quality. Recommended for most users. \|
	\| `stable-code-3b-Q4_K_S.gguf` \| 1.51 GB \| ~3.0 GB \| `Q4_K_S` \| ⭐⭐⭐½ \| Good speed/size balance, slight quality loss. \|
	\| `stable-code-3b-Q5_K_M.gguf` \| 1.86 GB \| ~3.4 GB \| `Q5_K_M` \| ⭐⭐⭐⭐½ \| Better quality than Q4, slightly larger. Great if you have the RAM. \|
	\| `stable-code-3b-Q5_K_S.gguf` \| 1.81 GB \| ~3.3 GB \| `Q5_K_S` \| ⭐⭐⭐⭐ \| Large but accurate. \|
	\| `stable-code-3b-Q6_K.gguf` \| 2.14 GB \| ~3.6 GB \| `Q6_K` \| ⭐⭐⭐⭐⭐ \| Near-perfect quality, very large. \|
	\| `stable-code-3b-Q8_0.gguf` \| 2.77 GB \| ~4.3 GB \| `Q8_0` \| ⭐⭐⭐⭐⭐ \| Closest to original quality. Use when RAM is not a concern. \|


	### 💡 Which file should I download?

	- Most users: `stable-code-3b-Q4_K_M.gguf` — best balance of size and quality
	- High RAM (32GB+): `stable-code-3b-Q8_0.gguf` — near-original quality
	- Low RAM (8GB): `stable-code-3b-Q3_K_M.gguf` — fits in 8GB with room to spare

	---

	## ⚡ Speed Benchmarks

	Run `python benchmark.py --model stable-code-3b` to generate speed results.


	---

	## 🧠 Quality Benchmarks

	Run `kaggle_bench.ipynb` on Kaggle to benchmark this model.


	---

	## 🚀 How to Use

	### Ollama
	```bash
	ollama run dhptl/stable-code-3b
	```

	### LM Studio / Jan / Open WebUI
	Search for `Dhptl/stable-code-3b` in the model browser.

	### llama.cpp CLI
	```bash
	# Download the binary from https://github.com/ggerganov/llama.cpp/releases
	./llama-cli \
	-m stable-code-3b-Q4_K_M.gguf \
	-p "You are a helpful assistant." \
	--conversation \
	-n 512
	```

	### Python — llama-cpp-python
	```python
	from llama_cpp import Llama

	llm = Llama(
	model_path="./stable-code-3b-Q4_K_M.gguf",
	n_gpu_layers=-1, # -1 = offload everything to GPU
	n_ctx=4096,
	)

	response = llm.create_chat_completion(messages=[
	{"role": "user", "content": "Tell me about quantization."}
	])
	print(response["choices"][0]["message"]["content"])
	```

	---

	## 🔍 About GGUF Quantization

	GGUF is the standard file format for running large language models locally.
	Quantization reduces the number of bits per weight:

	\| Format \| Bits/weight \| Size vs FP16 \| Quality \|
	\|---\|---\|---\|---\|
	\| Q2_K \| ~2.6 \| 16% \| ⭐ \|
	\| Q3_K_M \| ~3.3 \| 21% \| ⭐⭐⭐ \|
	\| Q4_K_M \| ~4.5 \| 28% \| ⭐⭐⭐⭐ ← sweet spot \|
	\| Q5_K_M \| ~5.6 \| 35% \| ⭐⭐⭐⭐½ \|
	\| Q8_0 \| ~8.5 \| 53% \| ⭐⭐⭐⭐⭐ \|

	---

	## 💬 Community & Feedback

	Found an issue? Have a question? Open a Discussion in the Community tab above.

	If these quantizations were useful, please consider:
	- ⭐ Starring [quant-kit](https://github.com/DhruvalPtl/quant-kit) on GitHub
	- 👍 Liking this model on HuggingFace
	- 💬 Leaving feedback in the Community tab