Instructions to use AeroxLabs/Aerova-Nova-2.5B-GGUF with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="AeroxLabs/Aerova-Nova-2.5B-GGUF")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModel
model = AutoModel.from_pretrained("AeroxLabs/Aerova-Nova-2.5B-GGUF", device_map="auto")

llama-cpp-python

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="AeroxLabs/Aerova-Nova-2.5B-GGUF",
	filename="aerova-nova-2b.gguf",
)

llm.create_chat_completion(
	messages = [
		{
			"role": "user",
			"content": "What is the capital of France?"
		}
	]
)

Notebooks
Google Colab
Kaggle
Local Apps Settings

llama.cpp

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with llama.cpp:

Install (macOS, Linux)

curl -LsSf https://llama.app/install.sh | sh
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf AeroxLabs/Aerova-Nova-2.5B-GGUF
# Run inference directly in the terminal:
llama cli -hf AeroxLabs/Aerova-Nova-2.5B-GGUF

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama serve -hf AeroxLabs/Aerova-Nova-2.5B-GGUF
# Run inference directly in the terminal:
llama cli -hf AeroxLabs/Aerova-Nova-2.5B-GGUF

Use pre-built binary

# Download pre-built binary from:
# https://github.com/ggerganov/llama.cpp/releases
# Start a local OpenAI-compatible server with a web UI:
./llama-server -hf AeroxLabs/Aerova-Nova-2.5B-GGUF
# Run inference directly in the terminal:
./llama-cli -hf AeroxLabs/Aerova-Nova-2.5B-GGUF

Build from source code

git clone https://github.com/ggerganov/llama.cpp.git
cd llama.cpp
cmake -B build
cmake --build build -j --target llama-server llama-cli
# Start a local OpenAI-compatible server with a web UI:
./build/bin/llama-server -hf AeroxLabs/Aerova-Nova-2.5B-GGUF
# Run inference directly in the terminal:
./build/bin/llama-cli -hf AeroxLabs/Aerova-Nova-2.5B-GGUF

Use Docker

docker model run hf.co/AeroxLabs/Aerova-Nova-2.5B-GGUF

LM Studio
Jan

vLLM

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "AeroxLabs/Aerova-Nova-2.5B-GGUF"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AeroxLabs/Aerova-Nova-2.5B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/AeroxLabs/Aerova-Nova-2.5B-GGUF

SGLang

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "AeroxLabs/Aerova-Nova-2.5B-GGUF" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AeroxLabs/Aerova-Nova-2.5B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "AeroxLabs/Aerova-Nova-2.5B-GGUF" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "AeroxLabs/Aerova-Nova-2.5B-GGUF",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with Ollama:
```
ollama run hf.co/AeroxLabs/Aerova-Nova-2.5B-GGUF
```
Unsloth Desktop
Docker Model Runner
How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with Docker Model Runner:
```
docker model run hf.co/AeroxLabs/Aerova-Nova-2.5B-GGUF
```

Lemonade

How to use AeroxLabs/Aerova-Nova-2.5B-GGUF with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull AeroxLabs/Aerova-Nova-2.5B-GGUF

Run and chat with the model

lemonade run user.Aerova-Nova-2.5B-GGUF-{{QUANT_TAG}}

List all available models

lemonade list

Atomic Chat

Aerova Nova

Medical intelligence, built for biomedical language.

Aerova Nova is a 7B-parameter biomedical language model developed by AeroxLabs, designed for medical, biological, and scientific text generation.

Built on BioMistral/BioMistral-7B-SLERP and trained with biomedical data from PubMed, Nova explores how open-source language models can be adapted toward specialized medical knowledge.

Nova is designed for research, experimentation, education, and biomedical NLP applications.

Model Overview


Model	Aerova Nova
Parameters	7B
Architecture	Causal Language Model
Base Model	BioMistral-7B-SLERP
Domain	Medicine & Biology
Languages	8
Training Data	NCBI PubMed
Format	GGUF
Inference	llama.cpp / llama-cpp-python
License	MIT

What Nova Can Do

Nova is designed for biomedical language tasks including:

Medical and biological text generation
Biomedical question answering
Scientific text completion
Medical knowledge exploration
Biomedical NLP research
Educational experimentation
Domain-specific language generation

Nova focuses on specialized biomedical language rather than general-purpose conversational intelligence.

Training

Nova builds upon BioMistral-7B-SLERP, adapting a strong open-source language model toward biomedical and medical language.

The primary dataset used in development is NCBI PubMed, providing large-scale biomedical literature for domain-specific language modeling.

Base model: BioMistral/BioMistral-7B-SLERP

Dataset: NCBI PubMed

Supported Languages

Nova's training configuration includes:

English
Spanish
German
Portuguese
Russian
French
Arabic
Chinese

Performance may vary between languages, particularly when working with specialized biomedical terminology.

Running Nova

Nova is distributed in GGUF format and is designed for efficient local inference with the llama.cpp ecosystem.

llama-cpp-python

Install the Python bindings:

pip install llama-cpp-python

Then load the model:

from llama_cpp import Llama

llm = Llama(
    model_path="./Aerova-Nova.gguf",
    n_ctx=4096,
    verbose=False,
)

output = llm(
    "Explain the role of hemoglobin in oxygen transport.",
    max_tokens=256,
    temperature=0.7,
)

print(output["choices"][0]["text"])

For compatible GPUs, layers can be offloaded with:

llm = Llama(
    model_path="./Aerova-Nova.gguf",
    n_ctx=4096,
    n_gpu_layers=-1,
)

For more advanced inference options, refer to the llama-cpp-python documentation.

Intended Use

Nova is intended for:

Biomedical research
Natural language processing research
Educational projects
Medical AI experimentation
Scientific text generation
Prototyping domain-specific AI systems
Local and private model inference

Nova can also serve as a foundation for further research and fine-tuning on specialized biomedical tasks.

Limitations & Safety

Nova is a research model and its outputs should be treated as generated information rather than authoritative medical advice.

The model may produce:

Incorrect or incomplete information
Hallucinated medical facts
Outdated information
Biased responses
Overconfident statements
Incorrect interpretations of scientific literature

Nova should not be used as a substitute for qualified medical professionals or as the sole basis for diagnosis, treatment, or other clinical decisions.

Always verify medical information using appropriate authoritative sources and qualified professionals.

Open Source

Aerova Nova is released openly so researchers, developers, and students can experiment with biomedical language models and build upon the work.

Open models make it possible to study specialized AI systems, reproduce experiments, and develop new applications without relying entirely on closed systems.

Nova is part of Aerova's open model research.