Instructions to use Aryanne/Zephyr-3.43B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use Aryanne/Zephyr-3.43B with Transformers:

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

pipe = pipeline("text-generation", model="Aryanne/Zephyr-3.43B", trust_remote_code=True)

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("Aryanne/Zephyr-3.43B", trust_remote_code=True, device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use Aryanne/Zephyr-3.43B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "Aryanne/Zephyr-3.43B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "Aryanne/Zephyr-3.43B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/Aryanne/Zephyr-3.43B

SGLang

How to use Aryanne/Zephyr-3.43B 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 "Aryanne/Zephyr-3.43B" \
    --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": "Aryanne/Zephyr-3.43B",
		"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 "Aryanne/Zephyr-3.43B" \
        --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": "Aryanne/Zephyr-3.43B",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use Aryanne/Zephyr-3.43B with Docker Model Runner:
```
docker model run hf.co/Aryanne/Zephyr-3.43B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

This model is a merge/fusion of Aryanne/Astridboros-3B and stabilityai/stablelm-zephyr-3b , 28 layers of Zephyr + 12 layers of Astridboros together(see zephyr-3.43b.yml or below).

A total of 40 layers, with 3.43B of parameters.

License it's the same as Zephyr cause it has 70% of it.

slices:
  - sources:
    - model: stabilityai/stablelm-zephyr-3b
      layer_range: [0, 14] 
  - sources:
    - model: Aryanne/Astridboros-3B
      layer_range: [10, 22]
  - sources:
    - model: stabilityai/stablelm-zephyr-3b
      layer_range: [18, 32]
merge_method: passthrough
dtype: float16

I recommend the use of the Zephyr prompt format.

<|user|>
List 3 synonyms for the word "tiny"<|endoftext|>
<|assistant|>
1. Dwarf
2. Little
3. Petite<|endoftext|>

GGUF Quants: afrideva/Zephyr-3.43B-GGUF A huge thanks to afrideva for the help. 🙌🤗

Downloads last month: 19

Safetensors

Model size

3B params

Tensor type

F16

Model tree for Aryanne/Zephyr-3.43B

Quantizations

1 model