---
license: gemma
finetune:
- google/medgemma-27b-it
tags:
- medical
- Gemma3
pipeline_tag: image-text-to-text
library_name: transformers
base_model:
- google/medgemma-27b-it
---
### About this model
This was created purely as an attempt to turn MedGemma-27B into a thinking and reasoning model. 
I will share it with everyone, so I would appreciate your feedback.

###Model Source
1. Merged publicly available Gemma3 models with outstanding scientific reasoning.
2. google/medgemma-27b-it

###What the model can do
1. Medical reasoning
2. Simple scientific reasoning
3. Simple coding and calculations
4. Tool calls

###Why is there no safetensors file?
To be honest, I created this using only what I could currently do at home, so I wanted some feedback. 
If it proves popular, I'll upload it, so please create a discussion.

```yaml
models:
  - model: gemma3-thinking(Original blend)
    parameters:
      weight: 1.0
  - model: google/medgemma-27b-it
    parameters: 
      weight: 1.618033988749
merge_method: nuslerp
tokenizer_source: google/medgemma-27b-it
dtype: bfloat16

```