MedGemma 1.5 4B MedCalcCaller (4-bit MLX-VLM)

This is a 4-bit MLX-VLM conversion of: sigjhl/medgemma-1.5-4b-it-MedCalcCaller

Conversion settings:

Tool: mlx_vlm.convert
Quantization: 4-bit
Group size: 64

This package preserves multimodal/vision components (Gemma3ForConditionalGeneration) and includes processor_config.json + preprocessor_config.json.

Downloads last month: 98

Safetensors

Model size

1B params

Tensor type

BF16

U32

MLX

Hardware compatibility

4-bit

Inference Providers NEW

This model isn't deployed by any Inference Provider. 🙋 Ask for provider support

Model tree for sigjhl/medgemma-1.5-4b-it-MedCalcCaller-4bit-mlx

Base model

google/medgemma-1.5-4b-it

Finetuned

sigjhl/medgemma-1.5-4b-it-MedCalcCaller

Quantized

(1)

this model