Instructions to use RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse with Transformers:

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

pipe = pipeline("text-generation", model="RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse")

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse")
model = AutoModelForCausalLM.from_pretrained("RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse", device_map="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Use Docker

docker model run hf.co/RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse

SGLang

How to use RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse 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 "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse" \
    --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": "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse",
		"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 "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse" \
        --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": "RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse",
		"prompt": "Once upon a time,",
		"max_tokens": 512,
		"temperature": 0.5
	}'

Docker Model Runner
How to use RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse with Docker Model Runner:
```
docker model run hf.co/RedHatAI/Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse
```

Llama-2-7b-evol-code-alpaca-pruned_70-quantized-deepsparse / README.md

Abhinav Agarwalla

Updating pruned70+quantized model

3f690cf over 2 years ago

preview code

Raw

History Blame Contribute Delete

2.46 kB

	---
	base_model: neuralmagic/Llama-2-7b-pruned70-retrained-evolcodealpaca
	inference: false
	model_type: llama
	pipeline_tag: text-generation
	datasets:
	- cerebras/SlimPajama-627B
	- theblackcat102/evol-codealpaca-v1
	tags:
	- sparse
	- code
	- deepsparse
	---

	# Llama-2-7b-pruned70-retrained-evolcodealpaca-quant-ds

	This repo contains a [70% sparse Llama 2 7B](https://huggingface.co/neuralmagic/Llama-2-7b-pruned70-retrained-evolcodealpaca) finetuned for code generation tasks using the [Evolved CodeAlpaca](https://huggingface.co/datasets/theblackcat102/evol-codealpaca-v1) dataset.
	It was then quantized to 8-bit weights + activations and exported to deploy with [DeepSparse](https://github.com/neuralmagic/deepsparse), a CPU inference runtime for sparse models.

	Official model weights from [Enabling High-Sparsity Foundational Llama Models with Efficient Pretraining and Deployment](https://arxiv.org/abs/2405.03594).

	Authors: Neural Magic, Cerebras

	## Usage

	Below we share some code snippets on how to get quickly started with running the model.

	### Sparse Transfer

	By leveraging a pre-sparsified model's structure, you can efficiently fine-tune on new data, leading to reduced hyperparameter tuning, training times, and computational costs. Learn about this process [here](https://neuralmagic.github.io/docs-v2/get-started/transfer).

	### Running the model

	For accelerated inference with sparsity on CPUs, deploy with [deepsparse](https://github.com/neuralmagic/deepsparse).

	```python
	# pip install deepsparse[llm]
	from deepsparse import TextGeneration

	model = TextGeneration(model_path="hf:neuralmagic/Llama-2-7b-pruned70-retrained-evolcodealpaca-quant-ds")

	input_text = "def fibonacci(n):\n"
	outputs = model(input_text, max_new_tokens=100)
	print(outputs.generations[0].text)
	```

	## Evaluation Benchmark Results

	Model evaluation metrics and results.

	\| Benchmark \| Metric \| Llama-2-7b-evolcodealpaca \| Llama-2-7b-pruned70-retrained-evolcodealpaca-quant-ds \|
	\|------------------------------------------------\|---------------\|-------------\|-------------------------------\|
	\| [HumanEval](https://arxiv.org/abs/2107.03374) \| pass@1 \| 32.03 \| 35.02 \|

	## Help

	For further support, and discussions on these models and AI in general, join [Neural Magic's Slack Community](https://join.slack.com/t/discuss-neuralmagic/shared_invite/zt-q1a1cnvo-YBoICSIw3L1dmQpjBeDurQ)