from deepsparse import TextGeneration

prompt = "How to make banana bread?"
formatted_prompt =  f"### User:\n{prompt}\n\n### Assistant:\n"

model = TextGeneration(model_path="hf:neuralmagic/Nous-Hermes-2-SOLAR-10.7B-pruned50-quant-ds")

print(model(formatted_prompt, max_new_tokens=200).generations[0].text)
"""
To make banana bread, you will need the following ingredients:

- 3 ripe bananas
- 1 cup of milk
- 1 cup of sugar
- 1/2 cup of butter
- 2 eggs
- 1 teaspoon of baking powder
- 1 teaspoon of salt
- 2 cups of flour

Here's a simple recipe to make banana bread:

1. Preheat your oven to 350°F (175°C).

2. In a large bowl, mash the ripe bananas.

3. Add the milk, sugar, butter, eggs, baking powder, salt, and flour to the mashed bananas. Mix everything together until you have a smooth batter.

4. Pour the batter into a greased loaf pan.

5. Bake the banana bread for about 60 minutes or until a tooth
"""

Prompt template


  ### User:\n
  {prompt}
  ### Assistant:\n

Sparsification

For details on how this model was sparsified, see the recipe.yaml in this repo and follow the instructions below.

bash

git clone https://github.com/neuralmagic/sparseml
pip install -e "sparseml[transformers]"
python sparseml/src/sparseml/transformers/sparsification/obcq/obcq.py NousResearch/Nous-Hermes-2-SOLAR-10.7B open_platypus --recipe recipe.yaml --save True
python sparseml/src/sparseml/transformers/sparsification/obcq/export.py --task text-generation --model_path obcq_deployment 
cp deployment/model.onnx deployment/model-orig.onnx

Run this kv-cache injection to speed up the model at inference by caching the Key and Value states:

python

import os
import onnx
from sparseml.exporters.kv_cache_injector import KeyValueCacheInjector
input_file = "deployment/model-orig.onnx"
output_file = "deployment/model.onnx"
model = onnx.load(input_file, load_external_data=False)
model = KeyValueCacheInjector(model_path=os.path.dirname(input_file)).apply(model)
onnx.save(model, output_file)
print(f"Modified model saved to: {output_file}")

Follow the instructions on our One Shot With SparseML page for a step-by-step guide for performing one-shot quantization of large language models.

Slack

For further support, and discussions on these models and AI in general, join Neural Magic's Slack Community

RedHatAI/Nous-Hermes-2-SOLAR-10.7B-pruned50-quant-ds

Nous Hermes 2 - Solar 10.7B - DeepSparse

Inference

Prompt template

Sparsification

Slack

RedHatAI/
Nous-Hermes-2-SOLAR-10.7B-pruned50-quant-ds