Instructions to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with Transformers:

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

pipe = pipeline("text-generation", model="adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning")
model = AutoModelForCausalLM.from_pretrained("adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning")
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]:]))

llama-cpp-python

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with llama-cpp-python:

# !pip install llama-cpp-python

from llama_cpp import Llama

llm = Llama.from_pretrained(
	repo_id="adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning",
	filename="ALLaM-7B-Instruct-preview-reasoning-Q8_0.gguf",
)

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

Notebooks
Google Colab
Kaggle
Local Apps

llama.cpp

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with llama.cpp:

Install from brew

brew install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
# Run inference directly in the terminal:
llama-cli -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

Install from WinGet (Windows)

winget install llama.cpp
# Start a local OpenAI-compatible server with a web UI:
llama-server -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
# Run inference directly in the terminal:
llama-cli -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

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 adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
# Run inference directly in the terminal:
./llama-cli -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

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 adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
# Run inference directly in the terminal:
./build/bin/llama-cli -hf adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

Use Docker

docker model run hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

LM Studio
Jan

vLLM

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

SGLang

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning 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 "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning" \
    --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": "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning",
		"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 "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning" \
        --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": "adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Ollama
How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with Ollama:
```
ollama run hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
```

Unsloth Studio new

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with Unsloth Studio:

Install Unsloth Studio (macOS, Linux, WSL)

curl -fsSL https://unsloth.ai/install.sh | sh
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning to start chatting

Install Unsloth Studio (Windows)

irm https://unsloth.ai/install.ps1 | iex
# Run unsloth studio
unsloth studio -H 0.0.0.0 -p 8888
# Then open http://localhost:8888 in your browser
# Search for adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning to start chatting

Using HuggingFace Spaces for Unsloth

# No setup required
# Open https://huggingface.co/spaces/unsloth/studio in your browser
# Search for adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning to start chatting

Docker Model Runner
How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with Docker Model Runner:
```
docker model run hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0
```

Lemonade

How to use adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning with Lemonade:

Pull the model

# Download Lemonade from https://lemonade-server.ai/
lemonade pull adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:Q8_0

Run and chat with the model

lemonade run user.ALLaM-7B-Instruct-preview-reasoning-Q8_0

List all available models

lemonade list

Disclaimer

This model is currently in the research phase and is not a final production-ready version. It is undergoing continuous testing, validation, and refinement. Performance, accuracy, and stability may vary, and results should not be considered definitive. Use at your own discretion. I do not guarantee the reliability of outputs, and any implementation or decision-making based on this model should be done with caution. Future updates may introduce significant changes or improvements. For inquiries or feedback, please reach out to me at Linkedin.

Usages:

install ollama then run ollama run hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning

from openai import OpenAI

client = OpenAI(
    base_url='http://localhost:11434/v1/',

    # required but ignored
    api_key='ollama',
)

# Create a chat completion request to the Ollama server.
response = client.chat.completions.create(
    model="hf.co/adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning:latest",  # Specify the model available on your Ollama server.
    messages=[
        {"role": "system", "content": """أجب بالصيغة التالية:
<think>
خطوات ايجاد الحل هنا
</think>
<answer>
الإجابة النهائية هنا
</answer>"""},
        {"role": "user", "content": """أوجد قيمة x في المعادلة التالية:
         2x + 3 = 7"""}
    ],
        max_tokens=4096,
        temperature=0.4,
        top_p=0.95,
)

# Output the response from the AI.
print(response.choices[0].message.content)
# Output:
"""
<think>
لحل المعادلة، نتبع الخطوات التالية:
1. نطرح 3 من كلا الجانبين للتخلص من العدد الثابت على الجانب الأيسر:
   2x + 3 - 3 = 7 - 3
   2x = 4

2. نقسم كلا الجانبين على 2 لحل x:
   (2x)/2 = 4/2
   x = 2
</think>
<answer>
قيمة x هي 2.
</answer>
"""

example:

Downloads last month: 81

Safetensors

Model size

7B params

Tensor type

BF16

Model tree for adeebaldkheel/ALLaM-7B-Instruct-preview-reasoning

Base model

humain-ai/ALLaM-7B-Instruct-preview

Quantized

(9)

this model

Quantizations

3 models