Instructions to use BinSaqban/Averroes-Q-Instruct-human55k with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use BinSaqban/Averroes-Q-Instruct-human55k with MLX:

# Make sure mlx-lm is installed
# pip install --upgrade mlx-lm

# Generate text with mlx-lm
from mlx_lm import load, generate

model, tokenizer = load("BinSaqban/Averroes-Q-Instruct-human55k")

prompt = "Write a story about Einstein"
messages = [{"role": "user", "content": prompt}]
prompt = tokenizer.apply_chat_template(
    messages, add_generation_prompt=True
)

text = generate(model, tokenizer, prompt=prompt, verbose=True)

Notebooks
Google Colab
Kaggle
Local Apps Settings
LM Studio

How to use BinSaqban/Averroes-Q-Instruct-human55k with Pi:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "BinSaqban/Averroes-Q-Instruct-human55k"

Configure the model in Pi

# Install Pi:
npm install -g @mariozechner/pi-coding-agent
# Add to ~/.pi/agent/models.json:
{
  "providers": {
    "mlx-lm": {
      "baseUrl": "http://localhost:8080/v1",
      "api": "openai-completions",
      "apiKey": "none",
      "models": [
        {
          "id": "BinSaqban/Averroes-Q-Instruct-human55k"
        }
      ]
    }
  }
}

Run Pi

# Start Pi in your project directory:
pi

Hermes Agent new

How to use BinSaqban/Averroes-Q-Instruct-human55k with Hermes Agent:

Start the MLX server

# Install MLX LM:
uv tool install mlx-lm
# Start a local OpenAI-compatible server:
mlx_lm.server --model "BinSaqban/Averroes-Q-Instruct-human55k"

Configure Hermes

# Install Hermes:
curl -fsSL https://hermes-agent.nousresearch.com/install.sh | bash
hermes setup
# Point Hermes at the local server:
hermes config set model.provider custom
hermes config set model.base_url http://127.0.0.1:8080/v1
hermes config set model.default BinSaqban/Averroes-Q-Instruct-human55k

Run Hermes

hermes

MLX LM

How to use BinSaqban/Averroes-Q-Instruct-human55k with MLX LM:

Generate or start a chat session

# Install MLX LM
uv tool install mlx-lm
# Interactive chat REPL
mlx_lm.chat --model "BinSaqban/Averroes-Q-Instruct-human55k"

Run an OpenAI-compatible server

# Install MLX LM
uv tool install mlx-lm
# Start the server
mlx_lm.server --model "BinSaqban/Averroes-Q-Instruct-human55k"
# Calling the OpenAI-compatible server with curl
curl -X POST "http://localhost:8000/v1/chat/completions" \
   -H "Content-Type: application/json" \
   --data '{
     "model": "BinSaqban/Averroes-Q-Instruct-human55k",
     "messages": [
       {"role": "user", "content": "Hello"}
     ]
   }'

Averroes-Q-Instruct-human55k

Arabic instruction-tuned model on 55K human-curated data.

Base: Qwen2.5-7B-Instruct Data: 55,174 human-curated Arabic instruction-conversation pairs LoRA: rank=8, scale=0.5, lr=1e-5, iters=5000 Train Loss: 1.434 (final) Training: ~45 min on M2 Ultra

Why "human55k"?

Previous versions (v1-v4) failed due to synthetic data (6M) and wrong configs. This version uses correct data + correct base + correct LoRA config.

Usage

from mlx_lm import load, generate
model, tokenizer = load("BinSaqban/Averroes-Q-Instruct-human55k")
prompt = "<|im_start|>user\nWhat is the capital of Saudi Arabia?<|im_end|>\n<|im_start|>assistant\n"
resp = generate(model, tokenizer, prompt=prompt, max_tokens=200)
print(resp)

License

Apache 2.0

Built by Hayula AI — Arabic-first, open-source, local-first.

Downloads last month: 11