Instructions to use CyberNative-AI/Colibri_8b_v0.1 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CyberNative-AI/Colibri_8b_v0.1 with Transformers:

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

pipe = pipeline("text-generation", model="CyberNative-AI/Colibri_8b_v0.1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CyberNative-AI/Colibri_8b_v0.1")
model = AutoModelForCausalLM.from_pretrained("CyberNative-AI/Colibri_8b_v0.1")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use CyberNative-AI/Colibri_8b_v0.1 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CyberNative-AI/Colibri_8b_v0.1"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CyberNative-AI/Colibri_8b_v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CyberNative-AI/Colibri_8b_v0.1

SGLang

How to use CyberNative-AI/Colibri_8b_v0.1 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 "CyberNative-AI/Colibri_8b_v0.1" \
    --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": "CyberNative-AI/Colibri_8b_v0.1",
		"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 "CyberNative-AI/Colibri_8b_v0.1" \
        --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": "CyberNative-AI/Colibri_8b_v0.1",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CyberNative-AI/Colibri_8b_v0.1 with Docker Model Runner:
```
docker model run hf.co/CyberNative-AI/Colibri_8b_v0.1
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Colibri_8b_v0.1

Colibri: Conversational CyberSecurity Model

TRY colibri via huggingface Spaces!

GGUF Quant used in Spaces available here: CyberNative-AI/Colibri_8b_v0.1_q5_gguf

Colibri_8b_v0.1 is a conversational model for cybersecurity fine-tuned from the awesome dolphin-2.9-llama3-8b. (llama-3-8b -> cognitivecomputations/dolphin-2.9-llama3-8b -> CyberNative-AI/Colibri_8b_v0.1)

We derived our training dataset by creating Q/A pairs from a huge amount of cybersecurity related texts.

v0.1 trained for 3 epochs using around 35k Q/A pairs FFT on all parameters using a single A100 80gb for 3 hours.

This model is using ChatML prompt template format.

example:

<|im_start|>system
You are Colibri, an advanced cybersecurity AI assistant developed by CyberNative AI.<|im_end|>
<|im_start|>user
{prompt}<|im_end|>
<|im_start|>assistant

ILLEGAL OR UNETHICAL USE IS PROHIBITED!