|
--- |
|
library_name: pytorch |
|
license: llama2 |
|
pipeline_tag: text-generation |
|
tags: |
|
- llm |
|
- generative_ai |
|
- quantized |
|
- android |
|
|
|
--- |
|
|
|
![](https://qaihub-public-assets.s3.us-west-2.amazonaws.com/qai-hub-models/models/llama_v2_7b_chat_quantized/web-assets/model_demo.png) |
|
|
|
# Llama-v2-7B-Chat: Optimized for Mobile Deployment |
|
## State-of-the-art large language model useful on a variety of language understanding and generation tasks |
|
|
|
Llama 2 is a family of LLMs. The "Chat" at the end indicates that the model is optimized for chatbot-like dialogue. The model is quantized to w4a16(4-bit weights and 16-bit activations) and part of the model is quantized to w8a16(8-bit weights and 16-bit activations) making it suitable for on-device deployment. For Prompt and output length specified below, the time to first token is Llama-PromptProcessor-Quantized's latency and average time per addition token is Llama-TokenGenerator-KVCache-Quantized's latency. |
|
|
|
This model is an implementation of Llama-v2-7B-Chat found [here](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf). |
|
This repository provides scripts to run Llama-v2-7B-Chat on Qualcomm® devices. |
|
More details on model performance across various devices, can be found |
|
[here](https://aihub.qualcomm.com/models/llama_v2_7b_chat_quantized). |
|
|
|
|
|
### Model Details |
|
|
|
- **Model Type:** Text generation |
|
- **Model Stats:** |
|
- Number of parameters: 7B |
|
- Precision: w4a16 + w8a16 (few layers) |
|
- Model-1 (Prompt Processor): Llama-PromptProcessor-Quantized |
|
- Max context length: 1024 |
|
- Prompt processor model size: 3.6 GB |
|
- Prompt processor input: 1024 tokens |
|
- Prompt processor output: 1024 output tokens + KVCache for token generator |
|
- Model-2 (Token Generator): Llama-TokenGenerator-KVCache-Quantized |
|
- Token generator model size: 3.6 GB |
|
- Token generator input: 1 input token + past KVCache |
|
- Token generator output: 1 output token + KVCache for next iteration |
|
- Decoding length: 1024 (1 output token + 1023 from KVCache) |
|
- Use: Initiate conversation with prompt-processor and then token generator for subsequent iterations. |
|
|
|
## Deploying Llama 2 on-device |
|
|
|
Large Language Model (LLM) such as [Llama 2](https://llama.meta.com/llama2/) has the following complexities to deploy on-device: |
|
1. Model size is too large to fit in device memory for inference |
|
2. Multi-Head Attention (MHA) has large activations leading to fallback from accelerators |
|
3. High model load and inference time |
|
|
|
We can tackle the above constraints with the following steps: |
|
1. Quantize weights to reduce on-disk model size, e.g., int8 or int4 weights |
|
2. Quantize activations to reduce inference time memory pressure |
|
3. Graph transformations to reduce inference time memory pressure, e.g., Multi-Head to Split-Head Attention (MHA -> SHA) |
|
4. Graph transformations to convert or decompose operations into more accelerator friendly operations e.g. Linear to Conv |
|
5. For LLM with 7B or more parameters, above steps are still not good enough on mobile, |
|
hence we go one step further and split model into sub-parts. |
|
|
|
Here, we divide the model into 4 parts in order to |
|
1. Make model exportable with low memory usage |
|
2. Avoid inference time out-of-memory errors |
|
|
|
In order to export Llama 2, please ensure |
|
1. Host machine has >40GB memory (RAM+swap-space) |
|
2. If you don't have enough memory, export.py will dump instructions to increase swap space accordingly. |
|
|
|
## Sample output prompts generated on-device |
|
1. --prompt "what is gravity?" --max-output-tokens 30 |
|
~~~ |
|
-------- Response Summary -------- |
|
Prompt: what is gravity? |
|
Response: Hello! I'm here to help you answer your question. Gravity is a fundamental force of nature that affects the behavior of objects with mass |
|
~~~ |
|
|
|
2. --prompt "what is 2+3?" --max-output-tokens 30 |
|
~~~ |
|
-------- Response Summary -------- |
|
Prompt: what is 2+3? |
|
Response: Of course! I'm happy to help! The answer to 2+3 is 5. |
|
~~~ |
|
|
|
3. --prompt "could you please write code for fibonacci series in python?" --max-output-tokens 100 |
|
~~~ |
|
-------- Response Summary -------- |
|
Prompt: could you please write code for fibonacci series in python? |
|
Response: Of course! Here is an example of how you could implement the Fibonacci sequence in Python: |
|
``` |
|
def fibonacci(n): |
|
if n <= 1: |
|
return n |
|
else: |
|
return fibonacci(n-1) + fibonacci(n-2) |
|
``` |
|
You can test the function by calling it with different values of `n`, like this: |
|
``` |
|
print(fibonacci(5)) |
|
~~~ |
|
|
|
|
|
|
|
| Device | Chipset | Target Runtime | Inference Time (ms) | Peak Memory Range (MB) | Precision | Primary Compute Unit | Target Model |
|
| ---|---|---|---|---|---|---|---| |
|
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 97.732 ms | 71 - 72 MB | UINT16 | NPU | Llama2-TokenGenerator-KVCache-Quantized |
|
| Samsung Galaxy S23 Ultra (Android 13) | Snapdragon® 8 Gen 2 | QNN Model Library | 2020.745 ms | 11 - 12 MB | UINT16 | NPU | Llama2-PromptProcessor-Quantized |
|
|
|
|
|
|
|
## Installation |
|
|
|
This model can be installed as a Python package via pip. |
|
|
|
```bash |
|
pip install "qai-hub-models[llama_v2_7b_chat_quantized]" |
|
``` |
|
|
|
|
|
|
|
## Configure Qualcomm® AI Hub to run this model on a cloud-hosted device |
|
|
|
Sign-in to [Qualcomm® AI Hub](https://app.aihub.qualcomm.com/) with your |
|
Qualcomm® ID. Once signed in navigate to `Account -> Settings -> API Token`. |
|
|
|
With this API token, you can configure your client to run models on the cloud |
|
hosted devices. |
|
```bash |
|
qai-hub configure --api_token API_TOKEN |
|
``` |
|
Navigate to [docs](https://app.aihub.qualcomm.com/docs/) for more information. |
|
|
|
|
|
|
|
## Demo off target |
|
|
|
The package contains a simple end-to-end demo that downloads pre-trained |
|
weights and runs this model on a sample input. |
|
|
|
```bash |
|
python -m qai_hub_models.models.llama_v2_7b_chat_quantized.demo |
|
``` |
|
|
|
The above demo runs a reference implementation of pre-processing, model |
|
inference, and post processing. |
|
|
|
**NOTE**: If you want running in a Jupyter Notebook or Google Colab like |
|
environment, please add the following to your cell (instead of the above). |
|
``` |
|
%run -m qai_hub_models.models.llama_v2_7b_chat_quantized.demo |
|
``` |
|
|
|
|
|
### Run model on a cloud-hosted device |
|
|
|
In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® |
|
device. This script does the following: |
|
* Performance check on-device on a cloud-hosted device |
|
* Downloads compiled assets that can be deployed on-device for Android. |
|
* Accuracy check between PyTorch and on-device outputs. |
|
|
|
```bash |
|
python -m qai_hub_models.models.llama_v2_7b_chat_quantized.export |
|
``` |
|
|
|
``` |
|
Profile Job summary of Llama2-TokenGenerator-KVCache-Quantized |
|
-------------------------------------------------- |
|
Device: Snapdragon X Elite CRD (11) |
|
Estimated Inference Time: 95.96 ms |
|
Estimated Peak Memory Range: 65.07-65.07 MB |
|
Compute Units: NPU (33818) | Total (33818) |
|
|
|
Profile Job summary of Llama2-PromptProcessor-Quantized |
|
-------------------------------------------------- |
|
Device: Snapdragon X Elite CRD (11) |
|
Estimated Inference Time: 1889.09 ms |
|
Estimated Peak Memory Range: 10.29-10.29 MB |
|
Compute Units: NPU (31766) | Total (31766) |
|
|
|
|
|
``` |
|
|
|
|
|
|
|
|
|
|
|
## Deploying compiled model to Android |
|
|
|
|
|
The models can be deployed using multiple runtimes: |
|
- TensorFlow Lite (`.tflite` export): [This |
|
tutorial](https://www.tensorflow.org/lite/android/quickstart) provides a |
|
guide to deploy the .tflite model in an Android application. |
|
|
|
|
|
- QNN (`.so` export ): This [sample |
|
app](https://docs.qualcomm.com/bundle/publicresource/topics/80-63442-50/sample_app.html) |
|
provides instructions on how to use the `.so` shared library in an Android application. |
|
|
|
|
|
## View on Qualcomm® AI Hub |
|
Get more details on Llama-v2-7B-Chat's performance across various devices [here](https://aihub.qualcomm.com/models/llama_v2_7b_chat_quantized). |
|
Explore all available models on [Qualcomm® AI Hub](https://aihub.qualcomm.com/) |
|
|
|
## License |
|
- The license for the original implementation of Llama-v2-7B-Chat can be found |
|
[here](https://github.com/facebookresearch/llama/blob/main/LICENSE). |
|
- The license for the compiled assets for on-device deployment can be found [here](https://github.com/facebookresearch/llama/blob/main/LICENSE) |
|
|
|
## References |
|
* [LLaMA: Open and Efficient Foundation Language Models](https://arxiv.org/abs/2302.13971) |
|
* [Source Model Implementation](https://huggingface.co/meta-llama/Llama-2-7b-chat-hf) |
|
|
|
## Community |
|
* Join [our AI Hub Slack community](https://aihub.qualcomm.com/community/slack) to collaborate, post questions and learn more about on-device AI. |
|
* For questions or feedback please [reach out to us](mailto:[email protected]). |
|
|
|
|
|
|