Downloads · 30 days
8
16% of all-time downloads
lumkomeds-gemma4-e4b
LeroyNgoma/lumkomeds-gemma4-e4b
lumkomeds-gemma4-e4b is a machine learning model from LeroyNgoma. Use it for the machine learning task on the model card, and read the license before you ship it in a product. The card lists the license as gemma.
Fine-tuned Gemma-4-E4B for clinical reasoning with Chain-of-Thought and tool calling.
All-time downloads
51
Public
Parameters
7.9B
15.9 GB on disk
Likes
0
Public
Repo makeup
Click a slice to open those files.
.safetensors15.9 GB · 100%
At a glance
- License
- gemma
- Model type
- gemma4
- Access
- Public
- Created
- Jun 23, 2026
- Updated
- Jun 23, 2026
- SHA
- 97027524
Model card
From the Hugging Face model README
LumkoMeds Clinical SLM (Gemma-4-E4B Fine-tuned)
Fine-tuned Gemma-4-E4B for clinical reasoning with Chain-of-Thought and tool calling.
Training
- Base: google/gemma-4-e4b
- Method: LoRA (rank 64, alpha 128, BF16) + DPO (beta 0.1)
- Strategy: DDP on 8x A100 80GB (data parallelism)
- Data: LumkoMeds gold v2 (41,279 SFT + 14,000 DPO examples)
- CoT: Chain-of-thought reasoning with <thinking> blocks
- Tools: sep_lookup, ddi_check, dosing_calculator, allergy_cross_check, formulary_check
- Safety: Refusal/boundary training curriculum
- Custom tokens: 11 non-splittable tokens for tool calls and clinical annotations
- Chat template: Custom with 'tool' role support
- Max sequence length: 8192
Usage
from transformers import AutoModelForCausalLM, AutoTokenizer
model = AutoModelForCausalLM.from_pretrained("path/to/merged_model")
tokenizer = AutoTokenizer.from_pretrained("path/to/merged_model")
- Type
- gemma4
- License
- gemma
- Created
- Jun 23, 2026
- Updated
- Jun 23, 2026