Instructions to use mlabonne/Zebrafish-7B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use mlabonne/Zebrafish-7B with Transformers:

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

pipe = pipeline("text-generation", model="mlabonne/Zebrafish-7B")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("mlabonne/Zebrafish-7B")
model = AutoModelForCausalLM.from_pretrained("mlabonne/Zebrafish-7B")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use mlabonne/Zebrafish-7B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "mlabonne/Zebrafish-7B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "mlabonne/Zebrafish-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/mlabonne/Zebrafish-7B

SGLang

How to use mlabonne/Zebrafish-7B 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 "mlabonne/Zebrafish-7B" \
    --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": "mlabonne/Zebrafish-7B",
		"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 "mlabonne/Zebrafish-7B" \
        --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": "mlabonne/Zebrafish-7B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use mlabonne/Zebrafish-7B with Docker Model Runner:
```
docker model run hf.co/mlabonne/Zebrafish-7B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Zebrafish-7B

Zebrafish-7B is my first model using the new merge method called Model Stock.

Zebrafish-7B is a merge of the following models using LazyMergekit:

Special thanks to Charles Goddard for the quick implementation!

🏆 Evaluation

Nous

Model	Average	AGIEval	GPT4All	TruthfulQA	Bigbench
mlabonne/AlphaMonarch-7B 📄	62.74	45.37	77.01	78.39	50.2
mlabonne/Zebrafish-7B 📄	62.41	44.92	77.18	78.25	49.28
mlabonne/Beyonder-4x7B-v3 📄	61.91	45.85	76.67	74.98	50.12
mlabonne/NeuralBeagle14-7B 📄	60.25	46.06	76.77	70.32	47.86
mistralai/Mistral-7B-Instruct-v0.2 📄	54.81	38.5	71.64	66.82	42.29

🧩 Configuration

models:
  - model: mistralai/Mistral-7B-v0.1
  - model: liminerity/M7-7b
  - model: rwitz/experiment26-truthy-iter-0
merge_method: model_stock
base_model: mistralai/Mistral-7B-v0.1
dtype: bfloat16

💻 Usage

!pip install -qU transformers accelerate

from transformers import AutoTokenizer
import transformers
import torch

model = "mlabonne/Zebrafish-7B"
messages = [{"role": "user", "content": "What is a large language model?"}]

tokenizer = AutoTokenizer.from_pretrained(model)
prompt = tokenizer.apply_chat_template(messages, tokenize=False, add_generation_prompt=True)
pipeline = transformers.pipeline(
    "text-generation",
    model=model,
    torch_dtype=torch.float16,
    device_map="auto",
)

outputs = pipeline(prompt, max_new_tokens=256, do_sample=True, temperature=0.7, top_k=50, top_p=0.95)
print(outputs[0]["generated_text"])