Instructions to use QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge with Transformers:

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

pipe = pipeline("text-generation", model="QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge")
model = AutoModelForCausalLM.from_pretrained("QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge")

Notebooks
Google Colab
Kaggle
Local Apps

vLLM

How to use QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge

SGLang

How to use QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge 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 "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge" \
    --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": "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge",
		"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 "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge" \
        --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": "QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge with Docker Model Runner:
```
docker model run hf.co/QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Details

This is an experimental merge I plan to use for future projects, it shows promising results from my limited testing. Further testing should probably be done! I just don't have the time, nor compute right now.

Configuration

The following YAML configuration was used to produce this model:

models:
  - model: unsloth/Llama-3.2-3B
    parameters:
      weight: 0.5
      density: 0.7

  - model: unsloth/Llama-3.2-3B-Instruct 
    parameters:
      weight: 0.5
      density: 0.6

merge_method: ties
base_model: unsloth/Llama-3.2-3B
parameters:
  normalize: true
  int8_mask: true
dtype: bfloat16
tokenizer_source: unsloth/Llama-3.2-3B

Downloads last month: 4

Safetensors

Model size

3B params

Tensor type

BF16

Model tree for QuietImpostor/Llama-3.2-3B-Instruct-Base-Merge

unsloth/Llama-3.2-3B

unsloth/Llama-3.2-3B-Instruct

Merge model

this model

Quantizations

1 model