What are the hardware requirements for Medfalcon 40B Lora?

Medfalcon 40B Lora requires approximately 0.9 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Medfalcon 40B Lora and how large is it?

Medfalcon 40B Lora is developed by nmitchko, a model with 40b 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 Medfalcon 40B Lora?

Medfalcon 40B Lora is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Medfalcon 40B Lora by nmitchko — VRAM 0.9GB

Name: Medfalcon 40B Lora
Author: nmitchko

Medfalcon 40B Lora is an open-source language model by nmitchko. Features: 40b LLM, VRAM: 0.9GB, License: cc-by-nc-3.0, LLM Explorer Score: 0.07.

Arxiv:2106.09685 Adapter En Finetuned Lora Medical Peft Region:us

Model Card on HF 🤗: https://huggingface.co/nmitchko/medfalcon-40b-lora

Medfalcon 40B Lora Parameters and Internals

Model Type

LoRA, text-generation

Use Cases

Areas:

medical domain

Applications:

question-answering and medical dialogue tasks

Additional Notes

This Lora supports 4-bit and 8-bit modes.

Training Details

Methodology:

LoRA, specifically QLora

Model Architecture:

Based on Falcon-40b-instruct

LLM Name	Medfalcon 40B Lora
Repository 🤗	https://huggingface.co/nmitchko/medfalcon-40b-lora
Model Size	40b
Required VRAM	0.9 GB
Updated	2026-07-14
Maintainer	nmitchko
Model Files	0.9 GB
Supported Languages	en
Model Architecture	Adapter
License	cc-by-nc-3.0
Is Biased	none
PEFT Type	LORA
LoRA Model	Yes
PEFT Target Modules	dense\|query_key_value\|dense_4h_to_h\|dense_h_to_4h
LoRA Alpha	16
LoRA Dropout	0.05
R Param	64

Rank the Medfalcon 40B Lora 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

Quantized Models of the Medfalcon 40B Lora

Model	Likes	Downloads	VRAM
Falcon 40B Instruct GPTQ	197	152	22 GB
Falcon 40B Instruct GGML	58	10	13 GB
Falcon 40B Instruct GPTQ	1	11	22 GB
Falcon 40B Instruct 8bit	6	12	41 GB

Best Alternatives to Medfalcon 40B Lora

Best Alternatives	Context / RAM	Downloads	Likes
Medfalcon V2 40B Lora	0K / 0.9 GB	7	3
Falcon 40B Code Alpaca Lora	0K / 0 GB	0	2
Falcon 40B Chat V0.1	0K / 0 GB	29	38
... 40B Int4 Peft Lora Sfttrainer	0K / 1.8 GB	0	12
Falcon 40B Guanaco Lora	0K / 0.2 GB	0	4
Falcon 40B Qlora Oasst De	0K / 1.8 GB	0	2
...lcon Lora 40b Ckpt 500 Oasst 1	0K / 1.8 GB	0	1
...e WizardLM Uncensored 40B Lora	0K / 0.1 GB	0	1

Note: green Score (e.g. "73.2") means that the model is better than nmitchko/medfalcon-40b-lora.