SOLAR-10.7B-Instruct-v1.0-pruned50-quant-ds

SOLAR-10.7B-Instruct-v1.0 - DeepSparse

This repo contains model files for SOLAR-10.7B-Instruct-v1.0 optimized for DeepSparse, a CPU inference runtime for sparse models.

This model was quantized and pruned with SparseGPT, using SparseML.

Inference

Install DeepSparse LLM for fast inference on CPUs:

pip install deepsparse-nightly[llm]

Run in a Python pipeline:

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/SOLAR-10.7B-Instruct-v1.0-pruned50-quant-ds")

print(model(formatted_prompt, max_new_tokens=200).generations[0].text)
"""
To make banana bread, follow these steps:

1. Gather ingredients:
- 4 ripe bananas
- 1 cup of flour (all-purpose)
- 1 teaspoon baking soda
- 1/2 cup of softened butter
- 1/2 cup of sugar
- 1/2 teaspoon salt
- 1 teaspoon vanilla extract
- 1/2 cup of milk

2. Preheat your oven: Preheat your oven to 350°F (177°C).

3. Prepare a loaf pan: Grease a loaf pan with butter or use a non-stick baking pan.

4. Mash the bananas: Peel the bananas and mash them in a bowl.

5. Mix the dry ingredients: In a separate bowl, mix the flour, baking soda, and salt.
"""

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.

git clone https://github.com/neuralmagic/sparseml
pip install -e "sparseml[transformers]"
python sparseml/src/sparseml/transformers/sparsification/obcq/obcq.py upstage/SOLAR-10.7B-Instruct-v1.0 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:

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