Instructions to use hayulalab/Averroes-Q-Instruct with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use hayulalab/Averroes-Q-Instruct with Transformers:

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

pipe = pipeline("text-generation", model="hayulalab/Averroes-Q-Instruct")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("hayulalab/Averroes-Q-Instruct")
model = AutoModelForCausalLM.from_pretrained("hayulalab/Averroes-Q-Instruct")
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 hayulalab/Averroes-Q-Instruct with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/hayulalab/Averroes-Q-Instruct

SGLang

How to use hayulalab/Averroes-Q-Instruct 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 "hayulalab/Averroes-Q-Instruct" \
    --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": "hayulalab/Averroes-Q-Instruct",
		"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 "hayulalab/Averroes-Q-Instruct" \
        --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": "hayulalab/Averroes-Q-Instruct",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use hayulalab/Averroes-Q-Instruct with Docker Model Runner:
```
docker model run hf.co/hayulalab/Averroes-Q-Instruct
```

hayulalab/Averroes-Q-Instruct

Averroes-Q-Instruct — Arabic-English bilingual instruction model, trained from scratch by Hayula Labs.

Training Details

Property	Value
Training	From scratch (random initialization)
Architecture	Decoder-only transformer (7B class)
Context	32,768 tokens
Data	Proprietary Hayula corpus — Arabic 40%, English 60%
Framework	MLX + Transformers

Capabilities

Native Arabic instruction following
English-Arabic bilingual
Long context (32K tokens)
Tool calling support

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model = AutoModelForCausalLM.from_pretrained("hayulalab/Averroes-Q-Instruct")
tokenizer = AutoTokenizer.from_pretrained("hayulalab/Averroes-Q-Instruct")

License

Research Only — No Commercial Use (CC BY 4.0).

Attribution required: "Based on hayulalab/Averroes-Q-Instruct by Hayula Labs"

Citation

@misc{hayulalab-averroes_q_instruct,
  author = {Hayula Labs},
  title = {hayulalab/Averroes-Q-Instruct},
  year = {2026},
  publisher = {HuggingFace},
  url = {https://huggingface.co/hayulalab/Averroes-Q-Instruct}
}

Built with ❤️ by Hayula Labs

Downloads last month: 779

Safetensors

Model size

8B params

Tensor type

BF16

Model tree for hayulalab/Averroes-Q-Instruct

Adapters

2 models