Literature · wanglab
Clinical Camel 70B
A 70-billion-parameter clinical language model fine-tuned for medical dialogue and clinical decision support.
What Clinical Camel 70B is
Clinical Camel 70B is a specialised clinical language model developed by wanglab. It is designed to support medical dialogue and clinical decision-making by combining broad linguistic capabilities with medical domain knowledge.
The model addresses the need for structured scientific reasoning in healthcare applications, including interpreting physician-patient interactions, processing clinical literature, and working through complex diagnostic pathways.
How it works
The system is built upon the Llama-2 70B architecture and fine-tuned using Quantised Low-Rank Adaptation (QLoRA), an efficient parameter-tuning method.
Training data comprises physician-patient dialogues, clinical articles, and MedQA-style reasoning chains, enabling the model to generate medically grounded text and multi-step reasoning.
Key facts
- Base model
- Llama-2 70B
- Parameter count
- 70 billion
- Tuning technique
- QLoRA
- Task
- Text generation
- Language
- English
What you can use it for
- Clinical decision support
- Generating multi-step reasoning chains for diagnostic and management queries grounded in clinical evidence.
- Medical dialogue generation
- Simulating and analysing conversational exchanges between patients and healthcare providers.
- Clinical question answering
- Resolving biomedical questions that require structured reasoning derived from MedQA-style knowledge.
Run it here
We list this model but cannot run it yet: excluded (gated): Gated and CC-BY-NC-4.0; a clinical LLM..
Everything on this page still applies. Running it needs a loader for this repository, not a bigger machine.
Strengths and limits
Good at
- Trained specifically on structured MedQA reasoning chains alongside clinical literature and patient interactions.
- Built upon a 70-billion-parameter foundation model adapted using memory-efficient QLoRA tuning.
Watch out for
- Operates across 70 billion parameters, requiring substantial hardware resources for local deployment and inference.
- Designed primarily for English-language medical dialogues and clinical texts.
Inputs and outputs
You provide
Prompt
The model continues from here.
You get back
The model's completion.
Licence
Check the licence before commercial use · cc-by-nc-4.0
This model is gated by its authors — its licence has to be accepted on the source repository before it will run.
Questions
Can I use Clinical Camel 70B commercially?
Check the licence before commercial use. This model is gated by its authors — its licence has to be accepted on the source repository before it will run.
What is the underlying model for Clinical Camel 70B?
The model is built on top of the Llama-2 70B architecture.
What data was used to fine-tune the model?
It was fine-tuned on physician-patient dialogues, clinical literature, and MedQA-style reasoning chains.
What tuning method was applied during training?
The developers used QLoRA to fine-tune the base model.
How to cite it
arXiv:2305.12031
Written from Clinical Camel 70B's own documentation (model card). Statements citing a figure that does not appear there are removed before publication. Licence and running details come from our registry.