Hugging Face's logo

qualcomm
/
Gemma-4-E2B-it

Text Generation
PyTorch
llm
vlm
generative_ai
android
Model card Files
xet
 Community

Gemma-4-E2B-it: Optimized for Qualcomm Devices

Gemma is a family of open models built by Google DeepMind. Gemma 4 models are multimodal, handling text and image input (with audio supported on small models) and generating text output. This release includes open-weights models in both pre-trained and instruction-tuned variants. Gemma 4 features a context window of up to 256K tokens and maintains multilingual support in over 140 languages.

This is based on the implementation of Gemma-4-E2B-it 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 Gemma-4-E2B-it on-device

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

Getting Started

There are two ways to deploy this model on your device:

Option 1: Download Pre-Exported Models

Below are pre-exported model assets ready for deployment.

Runtime Precision Chipset SDK Versions Download
GENIEX_LLAMACPP q4_0 Universal Download

For more device-specific assets and performance metrics, visit Gemma-4-E2B-it on Qualcomm® AI Hub.

Option 2: Export with Custom Configurations

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

This option is ideal if you need to customize the model beyond the default configuration provided here.

See our repository for Gemma-4-E2B-it on GitHub for usage instructions.

Model Details

Model Type: Model_use_case.text_generation

Model Stats:

  • Model architecture: Transformer with SigLIP-style ViT Vision Encoder, Per-Layer Embeddings (PLE), interleaved sliding-window/global attention with Grouped Query Attention (GQA), and shared KV layers.
  • Number of parameters: 2B (effective)
  • Response Rate: Rate of response generation after the first response token.
  • Supported languages: Multilingual (trained on 140+ languages)
  • 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.

Performance Summary

Model Runtime Precision Chipset Context Length Response Rate (tokens per second) Time To First Token (range, seconds)
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 512 31.705111 0.528357 - 2.113428
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 512 31.801764 0.28169533333333335 - 1.1267813333333334
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 512 33.695561 0.06813633333333333 - 0.2725453333333333
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 22.363562 1.2854541935483872 - 41.13453419354839
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 22.277421 0.4577602580645161 - 14.648328258064515
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite Gen 5 For Galaxy Mobile 4096 32.76263 0.07520329032258065 - 2.4065052903225808
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 512 34.222659 0.5953126666666667 - 2.381250666666667
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 512 29.879023 0.3048563333333333 - 1.2194253333333331
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 512 28.915732 0.07609099999999999 - 0.30436399999999997
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 4096 20.181043 1.411708193548387 - 45.174662193548386
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 4096 18.759262 0.5690559032258065 - 18.209788903225807
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® 8 Elite For Galaxy Mobile 4096 27.544753 0.09080200000000001 - 2.9056640000000002
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 512 53.469257 0.18787166666666666 - 0.7514866666666666
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 512 33.652612 0.15155066666666664 - 0.6062026666666666
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 512 30.331006 0.060608999999999996 - 0.24243599999999998
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 4096 43.359435 0.24944945161290324 - 7.982382451612904
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 4096 14.345144 0.17971787096774194 - 5.750971870967742
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X2 Elite 4096 14.99925 0.06681887096774194 - 2.1382038709677422
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 512 41.958422 0.4857046666666666 - 1.9428186666666665
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 512 29.585078 0.31198066666666663 - 1.2479226666666665
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 512 21.995727 0.14218133333333333 - 0.5687253333333333
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 4096 39.291585 0.45020680645161293 - 14.406617806451614
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 4096 18.449844 0.41507283870967737 - 13.282330838709676
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Snapdragon® X Elite 4096 20.882314 0.1356064193548387 - 4.339405419354838
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 512 10.463808 8.896726000000001 - 35.586904000000004
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 512 3.703586 13.634838 - 54.539352
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 512 14.707124 1.008742 - 4.034968
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 4096 6.88609 85.69991800000001 - 2742.3973760000003
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 4096 2.241468 158.02845499999998 - 5056.910559999999
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-8275 4096 12.612885 8.503915000000001 - 272.12528000000003
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 512 23.224758 3.750578 - 15.002312
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 512 10.010592 4.526033999999999 - 18.104135999999997
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 512 16.604906 0.941331 - 3.765324
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 512 41.958422 0.4857046666666666 - 1.9428186666666665
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 512 29.585078 0.31198066666666663 - 1.2479226666666665
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 512 21.995727 0.14218133333333333 - 0.5687253333333333
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 512 34.222659 0.5953126666666667 - 2.381250666666667
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 512 29.879023 0.3048563333333333 - 1.2194253333333331
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 512 28.915732 0.07609099999999999 - 0.30436399999999997
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 4096 18.118229 40.628958 - 1300.126656
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 4096 7.144893 54.815616 - 1754.099712
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-9075 4096 14.488949 8.055286 - 257.769152
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 4096 39.291585 0.45020680645161293 - 14.406617806451614
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 4096 18.449844 0.41507283870967737 - 13.282330838709676
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ IQ-X7181 4096 20.882314 0.1356064193548387 - 4.339405419354838
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 4096 20.181043 1.411708193548387 - 45.174662193548386
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 4096 18.759262 0.5690559032258065 - 18.209788903225807
Gemma-4-E2B-it GENIEX_LLAMACPP q4_0 Qualcomm® Dragonwing™ Q-8750 4096 27.544753 0.09080200000000001 - 2.9056640000000002

License

  • The license for the original implementation of Gemma-4-E2B-it 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