---
license: other
license_name: health-ai-developer-foundations
license_link: https://developers.google.com/health-ai-developer-foundations/terms
pipeline_tag: text-generation
extra_gated_heading: Access MedGemma on Hugging Face
extra_gated_prompt: >-
  To access MedGemma on Hugging Face, you're required to review and agree to
  [Health AI Developer Foundation's terms of
  use](https://developers.google.com/health-ai-developer-foundations/terms). To
  do this, please ensure you're logged in to Hugging Face and click below.
  Requests are processed immediately.
extra_gated_button_content: Acknowledge license
base_model: google/medgemma-27b-text-it
tags:
- medical
- clinical-reasoning
- thinking
---

# litert-community/MedGemma-27B-IT

This model provides a few variants of
[google/medgemma-27b-text-it](https://huggingface.co/google/medgemma-27b-text-it) that are ready for
deployment on Web using the 
[MediaPipe LLM Inference API](https://ai.google.dev/edge/mediapipe/solutions/genai/llm_inference).

### Web

* Build and run our [sample web app](https://github.com/google-ai-edge/mediapipe-samples/blob/main/examples/llm_inference/js/README.md).

To add the model to your web app, please follow the instructions in our [documentation](https://ai.google.dev/edge/mediapipe/solutions/genai/llm_inference/web_js).

## Performance

### Web

Note that all benchmark stats are from a MacBook Pro 2024 (Apple M4 Max chip) with 1280 KV cache size, 1024 tokens prefill, and 256 tokens decode, running in Chrome.

<table border="1">
  <tr>
   <th></th>
   <th>Precision</th>
   <th>Backend</th>
   <th>Prefill (tokens/sec)</th>
   <th>Decode (tokens/sec)</th>
   <th>Time-to-first-token (sec)</th>
   <th>GPU Memory</th>
   <th>CPU Memory</th>
   <th>Model size</th>
   <th></th>
  </tr>
  <tr>
  <td><p style="text-align: left">F16</p></td>
  <td><p style="text-align: left">int8</p></td>
  <td><p style="text-align: left">GPU</p></td>
  <td><p style="text-align: right">167 tk/s</p></td>
  <td><p style="text-align: right">8 tk/s</p></td>
  <td><p style="text-align: right">14.9 s</p></td>
  <td><p style="text-align: right">27.0 GB</p></td>
  <td><p style="text-align: right">1.5 GB</p></td>
  <td><p style="text-align: right">27.05 GB</p></td>
  <td><p style="text-align: left"><a style="text-decoration: none" href="https://huggingface.co/litert-community/MedGemma-27B-IT/resolve/main/medgemma-27b-it-int8-web.task">&#128279;</a></p></td>
  </tr>
  <td><p style="text-align: left">F32</p></td>
  <td><p style="text-align: left">int8</p></td>
  <td><p style="text-align: left">GPU</p></td>
  <td><p style="text-align: right">97 tk/s</p></td>
  <td><p style="text-align: right">8 tk/s</p></td>
  <td><p style="text-align: right">15.0 s</p></td>
  <td><p style="text-align: right">28.0 GB</p></td>
  <td><p style="text-align: right">1.5 GB</p></td>
  <td><p style="text-align: right">27.05 GB</p></td>
  <td><p style="text-align: left"><a style="text-decoration: none" href="https://huggingface.co/litert-community/MedGemma-27B-IT/resolve/main/medgemma-27b-it-int8-web.task">&#128279;</a></p></td>
  </tr>
</table>

*   Model size: measured by the size of the .tflite flatbuffer (serialization format for LiteRT models).
*   int8: quantized model with int8 weights and float activations.
*   GPU memory: measured by "GPU Process" memory for all of Chrome while running. Chrome was measured as using 340-350MB before any model loading took place.
*   CPU memory: measured for the entire tab while running. Tab was measured as using 60-70MB before any model loading took place.