Hugging Face's logo

qualcomm
/
PLaMo-1B

Text Generation
PyTorch
llm
generative_ai
quantized
android
Model card Files Community
PLaMo-1B / README.md
qaihm-bot's picture
qaihm-bot
Upload README.md with huggingface_hub
5c06652 verified
preview code
|
raw
history blame contribute delete
3.88 kB
metadata
library_name: pytorch
license: other
pipeline_tag: text-generation
tags:
  - llm
  - generative_ai
  - quantized
  - android

PLaMo-1B: Optimized for Mobile Deployment

State-of-the-art large language model useful on a variety of language understanding and generation tasks

PLaMo-1B is the first small language model (SLM) in the PLaMo™ Lite series from Preferred Networks (PFN), designed to power AI applications for edge devices including mobile, automotive, and robots across various industrial sectors. This model builds on the advancements of PLaMo-100B, a 100-billion parameter large language model (LLM) developed from the ground up by PFN’s subsidiary Preferred Elements (PFE). Leveraging high-quality Japanese and English text data generated by PLaMo-100B, PLaMo-1B has been pre-trained on a total of 4 trillion tokens. As a result, it delivers exceptional performance in Japanese benchmarks, outperforming other SLMs with similar parameter sizes. In evaluations such as Jaster 0-shot and 4-shot, PLaMo-1B has demonstrated performance on par with larger LLMs, making it a highly efficient solution for edge-based AI tasks.

More details on model performance accross various devices, can be found here.

Model Details

  • Model Type: Text generation
  • Model Stats:
    • Input sequence length for Prompt Processor: 128
    • Context length: 4096
    • Number of parameters: 1B
    • Precision: w4a16 + w8a16 (few layers)
    • Use: Initiate conversation with prompt-processor and then token generator for subsequent iterations.
    • Minimum QNN SDK version required: 2.27.7
    • Supported languages: Japanese and 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).
    • Response Rate: Rate of response generation after the first response token.
Model Device Chipset Target Runtime Response Rate (tokens per second) Time To First Token (range, seconds)
PLaMo-1B Snapdragon 8 Elite QRD Snapdragon® 8 Elite QNN 68.21 0.031448 - 1.006336

Deploying PLaMo-1B on-device

Please follow the LLM on-device deployment tutorial.

Community

Usage and Limitations

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