| --- |
| title: MicroLens |
| emoji: 🔬 |
| colorFrom: blue |
| colorTo: green |
| sdk: gradio |
| sdk_version: 5.7.1 |
| app_file: app.py |
| pinned: false |
| license: apache-2.0 |
| short_description: Gemma 4 E2B + LoRA · diatoms & fungal spores · 95 genera |
| models: |
| - Laborator/microlens-final |
| tags: |
| - microscopy |
| - biology |
| - vision-language |
| - gemma |
| - gemma-4 |
| - unsloth |
| - qlora |
| --- |
| |
| ## Demo video |
|
|
| [Watch the 90-second walkthrough on YouTube](https://youtu.be/r1GIi4EukVg) |
|
|
|
|
|
|
| # 🔬 MicroLens — Microscopy AI Demo |
|
|
| Live demo of **MicroLens**: a fine-tuned **Gemma 4 E2B** vision-language model that identifies microscopic subjects across **2 categories** (diatoms and fungal spores) and **95 genera**. |
|
|
| Built by **Serghei Brinza** (Vienna, Austria) for the **Kaggle Gemma 4 Good Hackathon** (May 2026). |
|
|
| --- |
|
|
| ## 🚀 How to use |
|
|
| 1. Upload a microscopy image (or pick one from the examples gallery). |
| 2. Optionally add a question, for example: "What is shown here?", "Describe the morphology", "What category does this belong to?" |
| 3. Click **AI Analyze**. |
|
|
| You get three side-by-side answers: |
|
|
| - **UNTRAINED BASELINE** — stock Gemma 4 E2B with no microscopy training. |
| - **MICROLENS · BRIEF** — same base + `Laborator/microlens-final` LoRA, prompted for a single-sentence genus answer. |
| - **MICROLENS · RICH** — same base + same LoRA, prompted for the full schema (genus + morphology + habitat + identification cues). |
|
|
| --- |
|
|
| ## 🔒 Privacy |
|
|
| - Images are processed in-memory on this Space and **not stored** server-side. |
| - **No analytics, no tracking, no third-party sharing.** |
| - For fully offline operation, see the desktop release linked below. |
|
|
| --- |
|
|
| ## ⚠️ Disclaimer |
|
|
| MicroLens is a **research and educational tool**. It is **not a medical device** and must not be used for clinical diagnosis, food-safety decisions, forensic identification, or any purpose where an incorrect answer could harm a person. |
|
|
| - Predictions are probabilistic and may be wrong, especially on out-of-distribution samples. |
| - Always verify critical findings with a qualified microscopist or laboratory. |
|
|
| For any vertical that requires regulatory clearance (medicine, veterinary, forensics, food safety), see the [4-stage gated rollout](https://github.com/SergheiBrinza/microlens/blob/main/ROADMAP.md) in the project roadmap. |
|
|
| --- |
|
|
| ## 🔗 Links |
|
|
| - **Model:** [Laborator/microlens-final](https://huggingface.co/Laborator/microlens-final) |
| - **Source code:** [SergheiBrinza/microlens](https://github.com/SergheiBrinza/microlens) |
| - **Hackathon writeup:** [KAGGLE_WRITEUP.md](https://github.com/SergheiBrinza/microlens/blob/main/KAGGLE_WRITEUP.md) |
| - **Roadmap:** [ROADMAP.md](https://github.com/SergheiBrinza/microlens/blob/main/ROADMAP.md) |
|
|
| --- |
|
|
| ## 📜 License and citation |
|
|
| - **Code:** Apache 2.0 |
| - **Documentation:** CC-BY 4.0 |
|
|
| ```bibtex |
| @software{brinza_microlens_2026, |
| title = {MicroLens: a microscopy vision-language interface built on Gemma 4 E2B}, |
| author = {Brinza, Serghei}, |
| year = {2026}, |
| month = may, |
| publisher = {Hugging Face}, |
| note = {Submission to the Gemma 4 Good Hackathon (Kaggle, May 2026)} |
| } |
| ``` |
|
|
| --- |
|
|
| Built with **Unsloth**, **llama.cpp**, and **Gradio** in Vienna, Austria. |
|
|
