Instructions to use bartowski/wavecoder-ultra-6.7b-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use bartowski/wavecoder-ultra-6.7b-GGUF with Transformers:

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

pipe = pipeline("text-generation", model="bartowski/wavecoder-ultra-6.7b-GGUF")

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("bartowski/wavecoder-ultra-6.7b-GGUF", dtype="auto")

llama-cpp-python

How to use bartowski/wavecoder-ultra-6.7b-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="bartowski/wavecoder-ultra-6.7b-GGUF",
	filename="wavecoder-ultra-6.7b-IQ1_M.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use bartowski/wavecoder-ultra-6.7b-GGUF with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf bartowski/wavecoder-ultra-6.7b-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 bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
# Run inference directly in the terminal:
llama-cli -hf bartowski/wavecoder-ultra-6.7b-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 bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
# Run inference directly in the terminal:
./llama-cli -hf bartowski/wavecoder-ultra-6.7b-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 bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
# Run inference directly in the terminal:
./build/bin/llama-cli -hf bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M

Use Docker

docker model run hf.co/bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M

LM Studio
Jan

vLLM

How to use bartowski/wavecoder-ultra-6.7b-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "bartowski/wavecoder-ultra-6.7b-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "bartowski/wavecoder-ultra-6.7b-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M

SGLang

How to use bartowski/wavecoder-ultra-6.7b-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 "bartowski/wavecoder-ultra-6.7b-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": "bartowski/wavecoder-ultra-6.7b-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 "bartowski/wavecoder-ultra-6.7b-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": "bartowski/wavecoder-ultra-6.7b-GGUF",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Ollama
How to use bartowski/wavecoder-ultra-6.7b-GGUF with Ollama:
```
ollama run hf.co/bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
```

Unsloth Studio new

How to use bartowski/wavecoder-ultra-6.7b-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 bartowski/wavecoder-ultra-6.7b-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 bartowski/wavecoder-ultra-6.7b-GGUF to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for bartowski/wavecoder-ultra-6.7b-GGUF to start chatting

Docker Model Runner
How to use bartowski/wavecoder-ultra-6.7b-GGUF with Docker Model Runner:
```
docker model run hf.co/bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M
```

Lemonade

How to use bartowski/wavecoder-ultra-6.7b-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull bartowski/wavecoder-ultra-6.7b-GGUF:Q4_K_M

Run and chat with the model

lemonade run user.wavecoder-ultra-6.7b-GGUF-Q4_K_M

List all available models

lemonade list

Llamacpp Quantizations of wavecoder-ultra-6.7b

Using llama.cpp release b2675 for quantization.

Original model: https://huggingface.co/microsoft/wavecoder-ultra-6.7b

All quants made using imatrix option with dataset provided by Kalomaze here

Prompt format

This seems to follow the DeepSeek coder format, aka Alpaca.

{system_prompt}
### Instruction: {prompt}
### Response:

Download a file (not the whole branch) from below:

Filename	Quant type	File Size	Description
wavecoder-ultra-6.7b-Q8_0.gguf	Q8_0	7.16GB	Extremely high quality, generally unneeded but max available quant.
wavecoder-ultra-6.7b-Q6_K.gguf	Q6_K	5.52GB	Very high quality, near perfect, recommended.
wavecoder-ultra-6.7b-Q5_K_M.gguf	Q5_K_M	4.78GB	High quality, recommended.
wavecoder-ultra-6.7b-Q5_K_S.gguf	Q5_K_S	4.65GB	High quality, recommended.
wavecoder-ultra-6.7b-Q4_K_M.gguf	Q4_K_M	4.08GB	Good quality, uses about 4.83 bits per weight, recommended.
wavecoder-ultra-6.7b-Q4_K_S.gguf	Q4_K_S	3.85GB	Slightly lower quality with more space savings, recommended.
wavecoder-ultra-6.7b-IQ4_NL.gguf	IQ4_NL	3.82GB	Decent quality, slightly smaller than Q4_K_S with similar performance recommended.
wavecoder-ultra-6.7b-IQ4_XS.gguf	IQ4_XS	3.62GB	Decent quality, smaller than Q4_K_S with similar performance, recommended.
wavecoder-ultra-6.7b-Q3_K_L.gguf	Q3_K_L	3.59GB	Lower quality but usable, good for low RAM availability.
wavecoder-ultra-6.7b-Q3_K_M.gguf	Q3_K_M	3.29GB	Even lower quality.
wavecoder-ultra-6.7b-IQ3_M.gguf	IQ3_M	3.11GB	Medium-low quality, new method with decent performance comparable to Q3_K_M.
wavecoder-ultra-6.7b-IQ3_S.gguf	IQ3_S	2.94GB	Lower quality, new method with decent performance, recommended over Q3_K_S quant, same size with better performance.
wavecoder-ultra-6.7b-Q3_K_S.gguf	Q3_K_S	2.94GB	Low quality, not recommended.
wavecoder-ultra-6.7b-IQ3_XS.gguf	IQ3_XS	2.79GB	Lower quality, new method with decent performance, slightly better than Q3_K_S.
wavecoder-ultra-6.7b-IQ3_XXS.gguf	IQ3_XXS	2.58GB	Lower quality, new method with decent performance, comparable to Q3 quants.
wavecoder-ultra-6.7b-Q2_K.gguf	Q2_K	2.53GB	Very low quality but surprisingly usable.
wavecoder-ultra-6.7b-IQ2_M.gguf	IQ2_M	2.36GB	Very low quality, uses SOTA techniques to also be surprisingly usable.
wavecoder-ultra-6.7b-IQ2_S.gguf	IQ2_S	2.19GB	Very low quality, uses SOTA techniques to be usable.
wavecoder-ultra-6.7b-IQ2_XS.gguf	IQ2_XS	2.03GB	Very low quality, uses SOTA techniques to be usable.
wavecoder-ultra-6.7b-IQ2_XXS.gguf	IQ2_XXS	1.85GB	Lower quality, uses SOTA techniques to be usable.
wavecoder-ultra-6.7b-IQ1_M.gguf	IQ1_M	1.65GB	Extremely low quality, not recommended.
wavecoder-ultra-6.7b-IQ1_S.gguf	IQ1_S	1.52GB	Extremely low quality, not recommended.

Which file should I choose?

A great write up with charts showing various performances is provided by Artefact2 here

The first thing to figure out is how big a model you can run. To do this, you'll need to figure out how much RAM and/or VRAM you have.

If you want your model running as FAST as possible, you'll want to fit the whole thing on your GPU's VRAM. Aim for a quant with a file size 1-2GB smaller than your GPU's total VRAM.

If you want the absolute maximum quality, add both your system RAM and your GPU's VRAM together, then similarly grab a quant with a file size 1-2GB Smaller than that total.

Next, you'll need to decide if you want to use an 'I-quant' or a 'K-quant'.

If you don't want to think too much, grab one of the K-quants. These are in format 'QX_K_X', like Q5_K_M.

If you want to get more into the weeds, you can check out this extremely useful feature chart:

llama.cpp feature matrix

But basically, if you're aiming for below Q4, and you're running cuBLAS (Nvidia) or rocBLAS (AMD), you should look towards the I-quants. These are in format IQX_X, like IQ3_M. These are newer and offer better performance for their size.

These I-quants can also be used on CPU and Apple Metal, but will be slower than their K-quant equivalent, so speed vs performance is a tradeoff you'll have to decide.

The I-quants are not compatible with Vulcan, which is also AMD, so if you have an AMD card double check if you're using the rocBLAS build or the Vulcan build. At the time of writing this, LM Studio has a preview with ROCm support, and other inference engines have specific builds for ROCm.

Want to support my work? Visit my ko-fi page here: https://ko-fi.com/bartowski

Downloads last month: 125

GGUF

Model size

7B params

Architecture

llama

Hardware compatibility

1-bit

2-bit

3-bit

4-bit

5-bit

6-bit

8-bit