Instructions to use BrelloES/Brello-QAI-0.6B with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BrelloES/Brello-QAI-0.6B with Transformers:

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

pipe = pipeline("text-generation", model="BrelloES/Brello-QAI-0.6B")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("BrelloES/Brello-QAI-0.6B")
model = AutoModelForCausalLM.from_pretrained("BrelloES/Brello-QAI-0.6B")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use BrelloES/Brello-QAI-0.6B with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "BrelloES/Brello-QAI-0.6B"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "BrelloES/Brello-QAI-0.6B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/BrelloES/Brello-QAI-0.6B

SGLang

How to use BrelloES/Brello-QAI-0.6B 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 "BrelloES/Brello-QAI-0.6B" \
    --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": "BrelloES/Brello-QAI-0.6B",
		"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 "BrelloES/Brello-QAI-0.6B" \
        --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": "BrelloES/Brello-QAI-0.6B",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use BrelloES/Brello-QAI-0.6B with Docker Model Runner:
```
docker model run hf.co/BrelloES/Brello-QAI-0.6B
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

Brello QAI 0.6B by Epic Systems

Brello QAI is a proprietary model created by Epic Systems. It is designed to deliver deep, targeted, and actionable responses rather than generic overviews. It supports both thinking and non-thinking modes.

Strong reasoning via Qwen3 thinking tokens (...)
Default Brello persona prioritizes specificity and actionable guidance
Lightweight local inference on macOS (MPS), CUDA, or CPU

Quickstart

Local run with the included script:

python3 -m venv .venv
source .venv/bin/activate
pip install --upgrade pip
pip install -r requirements.txt
python run.py --prompt "Help me design a rollout plan for feature flags in a mobile app."

Options:

--system: override the default Brello persona; pass empty string to disable
--no-think: disable thinking mode
--device: auto|mps|cuda|cpu

Example with a custom system message and non-thinking mode:

python run.py --no-think --system "You are Brello QAI. Give surgical, stepwise answers with tradeoffs." \
  --prompt "Compare blue/green and canary deployments for a small backend team."

Light SFT (optional)

To nudge the style further toward “deep and targeted,” see training/ for a LoRA SFT script and a small sample dataset. You can train a tiny adapter on your examples and then merge it into a standalone Brello checkpoint.

License

This repository includes open-source components under the Apache-2.0 license. See LICENSE for details.

Downloads last month: 2

Safetensors

Model size

0.6B params

Tensor type

F32

Model tree for BrelloES/Brello-QAI-0.6B

Quantizations

1 model