Hugging Face's logo

qualcomm
/
Llama-v3.2-3B-Instruct

Text Generation
PyTorch
llm
generative_ai
android
Model card Files
xet
 Community

Llama-v3.2-3B-Instruct: Optimized for Qualcomm Devices

Llama 3 is a family of LLMs. 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.

This is based on the implementation of Llama-v3.2-3B-Instruct found here. This repository contains pre-exported model files optimized for Qualcomm® devices. You can use the Qualcomm® AI Hub Models library to export with custom configurations. More details on model performance across various devices, can be found here.

Qualcomm AI Hub Models uses Qualcomm AI Hub Workbench to compile, profile, and evaluate this model. Sign up to run these models on a hosted Qualcomm® device.

Deploying Llama-v3.2-3B-Instruct on-device

Follow the GenieX quickstart to install GenieX and deploy the model on a target device.

You'll need to export the model artifact using the steps below, then follow Run a Local Model with GenieX.

See the LLM-on-Genie tutorial to run with the Genie runtime. Note: Genie support will be deprecated soon.

Getting Started

Due to licensing restrictions, we cannot distribute pre-exported model assets for this model. Use the Qualcomm® AI Hub Models Python library to compile and export the model with your own:

  • Custom weights (e.g., fine-tuned checkpoints)
  • Custom input shapes
  • Target device and runtime configurations

See our repository for Llama-v3.2-3B-Instruct on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.text_generation

Model Stats:

  • Response Rate: Rate of response generation after the first response token.
  • Supported languages: English.
  • TTFT: Time To First Token is the time it takes to generate the first response token. This is expressed as a range because it varies based on the length of the prompt. The lower bound is for a short prompt (up to 128 tokens, i.e., one iteration of the prompt processor) and the upper bound is for a prompt using the full context length (4096 tokens).

Performance Summary

Model Runtime Precision Chipset Context Length Response Rate (tokens per second) Time To First Token (range, seconds)
Llama-v3.2-3B-Instruct GENIE w4 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 15.56686782836914 0.126388 - 4.044416
Llama-v3.2-3B-Instruct GENIE w4 Snapdragon® 8 Elite For Galaxy Mobile 4096 14.32213306427002 0.141308 - 4.521856
Llama-v3.2-3B-Instruct GENIE w4 Snapdragon® X2 Elite 4096 15.719315528869629 0.13342500000000002 - 4.2696000000000005
Llama-v3.2-3B-Instruct GENIE w4 Snapdragon® X Elite 4096 5.476031303405762 0.257699 - 8.246368
Llama-v3.2-3B-Instruct GENIE w4 Qualcomm® Dragonwing™ IQ-8275 4096 7.600459098815918 0.397159 - 12.709088
Llama-v3.2-3B-Instruct GENIE w4 Qualcomm® Dragonwing™ IQ-9075 4096 8.157272338867188 0.311378 - 9.964096
Llama-v3.2-3B-Instruct GENIE w4 Qualcomm® Dragonwing™ IQ-X7181 4096 5.476031303405762 0.257699 - 8.246368
Llama-v3.2-3B-Instruct GENIE w4 Qualcomm® Dragonwing™ Q-8750 4096 14.32213306427002 0.141308 - 4.521856
Llama-v3.2-3B-Instruct GENIE w4a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 28.16663360595703 0.072227 - 2.311264
Llama-v3.2-3B-Instruct GENIE w4a16 Snapdragon® 8 Elite For Galaxy Mobile 4096 26.502702713012695 0.07698 - 2.46336
Llama-v3.2-3B-Instruct GENIE w4a16 Snapdragon® X2 Elite 4096 39.84222412109375 0.099425 - 3.1816
Llama-v3.2-3B-Instruct GENIE w4a16 Snapdragon® X Elite 4096 12.248597145080566 0.11766599999999999 - 3.7653119999999998
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® Dragonwing™ IQ-8275 4096 16.66777801513672 0.242084 - 7.746688
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® SA8650P 4096 17.47 0.109614 - 3.507648
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® SA8255P 4096 17.47 0.109614 - 3.507648
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® Dragonwing™ IQ-9075 4096 16.845792770385742 0.10630500000000001 - 3.4017600000000003
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® Dragonwing™ IQ-X7181 4096 12.248597145080566 0.11766599999999999 - 3.7653119999999998
Llama-v3.2-3B-Instruct GENIE w4a16 Qualcomm® Dragonwing™ Q-8750 4096 26.502702713012695 0.07698 - 2.46336
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 14.005415 0.22386209677419355 - 7.163587096774194
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Snapdragon® 8 Elite For Galaxy Mobile 4096 12.357958 0.16279796774193547 - 5.209534967741935
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Snapdragon® X2 Elite 4096 14.18382 0.17576290322580643 - 5.624412903225806
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Snapdragon® X Elite 4096 7.855231 0.24648012903225805 - 7.887364129032258
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Qualcomm® Dragonwing™ IQ-8275 4096 7.492126 0.3023875806451613 - 9.676402580645162
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Qualcomm® Dragonwing™ IQ-9075 4096 7.806524 0.28800796774193543 - 9.216254967741934
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Qualcomm® Dragonwing™ IQ-X7181 4096 7.855231 0.24648012903225805 - 7.887364129032258
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4 Qualcomm® Dragonwing™ Q-8750 4096 12.357958 0.16279796774193547 - 5.209534967741935
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 24.547539 0.079366 - 2.539712
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Snapdragon® 8 Elite For Galaxy Mobile 4096 22.230211 0.08288209677419354 - 2.6522270967741934
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Snapdragon® X2 Elite 4096 35.463076 0.061370548387096775 - 1.9638575483870968
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Snapdragon® X Elite 4096 18.592517 0.11378725806451614 - 3.6411922580645166
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Qualcomm® Dragonwing™ IQ-8275 4096 16.029866 0.11257732258064516 - 3.6024743225806453
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Qualcomm® Dragonwing™ IQ-9075 4096 16.852771 0.1103741935483871 - 3.531974193548387
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Qualcomm® Dragonwing™ IQ-X7181 4096 18.592517 0.11378725806451614 - 3.6411922580645166
Llama-v3.2-3B-Instruct GENIEX_QAIRT w4a16 Qualcomm® Dragonwing™ Q-8750 4096 22.230211 0.08288209677419354 - 2.6522270967741934

License

  • The license for the original implementation of Llama-v3.2-3B-Instruct can be found here.

References

Community

Usage and Limitations

This model may not be used for or in connection with any of the following applications:

  • Accessing essential private and public services and benefits;
  • Administration of justice and democratic processes;
  • Assessing or recognizing the emotional state of a person;
  • Biometric and biometrics-based systems, including categorization of persons based on sensitive characteristics;
  • Education and vocational training;
  • Employment and workers management;
  • Exploitation of the vulnerabilities of persons resulting in harmful behavior;
  • General purpose social scoring;
  • Law enforcement;
  • Management and operation of critical infrastructure;
  • Migration, asylum and border control management;
  • Predictive policing;
  • Real-time remote biometric identification in public spaces;
  • Recommender systems of social media platforms;
  • Scraping of facial images (from the internet or otherwise); and/or
  • Subliminal manipulation
Downloads last month

-

Downloads are not tracked for this model. How to track
Inference Providers NEW
Text Generation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support