What are the hardware requirements for Medra4b?

Medra4b requires approximately 8.6 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Medra4b and how large is it?

Medra4b is developed by drwlf, a model with 4b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Medra4b?

Medra4b is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Medra4b by drwlf — VRAM 8.6GB

Name: Medra4b
Author: drwlf

Medra4b is an open-source language model by drwlf. Features: 4b LLM, VRAM: 8.6GB, License: apache-2.0, LLM Explorer Score: 0.18.

Base model:finetune:google/gem... Base model:google/gemma-3-4b-i... Conversational Dataset:drwlf/medra-thinking-7... Diagnostic-reasoning En Finetuned Gemma-3 Gemma3 Medical-ai Region:us Ro Safetensors Sharded Summarization Tensorflow

Model Card on HF 🤗: https://huggingface.co/drwlf/Medra4b

Medra4b Parameters and Internals

LLM Name	Medra4b
Repository 🤗	https://huggingface.co/drwlf/Medra4b
Base Model(s)	Gemma 3 4B It google/gemma-3-4b-it
Model Size	4b
Required VRAM	8.6 GB
Updated	2026-08-08
Maintainer	drwlf
Model Type	gemma3
Model Files	5.0 GB: 1-of-2 3.6 GB: 2-of-2
Supported Languages	en ro
Model Architecture	Gemma3ForConditionalGeneration
License	apache-2.0
Transformers Version	4.52.4
Tokenizer Class	GemmaTokenizer
Padding Token	<pad>
Torch Data Type	bfloat16

Rank the Medra4b capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Medra4b

Best Alternatives	Context / RAM	Downloads	Likes
Medgemma 4B It Abliterated	128K / 6.3 GB	13	6
Gemma 3 Glitter 4B	128K / 8.6 GB	334	7
Fallen Gemma3 4B V1	128K / 8.6 GB	13	18
Gemma 3 4B It	0K / 8.6 GB	1855262	1345
Gemma 3 4B It	0K / 8.6 GB	62743	28
Translategemma 4B It	0K / 8.6 GB	38998	793
...rblock5 Gemma 3 4B It Seed1010	0K / 8.6 GB	84	1
...Gptmini Gemma 3 4B It Seed1010	0K / 8.6 GB	21	1
...block35 Gemma 3 4B It Seed1010	0K / 8.6 GB	17	1
Nemotron 3 Content Safety	0K / 8.6 GB	1706	18

Note: green Score (e.g. "73.2") means that the model is better than drwlf/Medra4b.