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

Libraries

How to use humain-ai/ALLaM-7B-Instruct-preview with Transformers:

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

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

# pip install -U transformers accelerate
# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("humain-ai/ALLaM-7B-Instruct-preview")
model = AutoModelForCausalLM.from_pretrained("humain-ai/ALLaM-7B-Instruct-preview", device_map="auto")
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=256)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use humain-ai/ALLaM-7B-Instruct-preview with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/humain-ai/ALLaM-7B-Instruct-preview

SGLang

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

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

Fix typo in evaluation scores link

by aalquwayfili - opened Nov 18, 2025

base: refs/heads/main

←

from: refs/pr/8

Discussion Files changed

-1

Files changed (1) hide show

README.md +1 -1

README.md CHANGED Viewed

@@ -128,7 +128,7 @@ ALLaM is a generative model that comes with inherent uncertainties. Trials canno
 All models were evaluated using our proprietary evaluation pipeline and [LM Evaluation Harness framework](https://github.com/EleutherAI/lm-evaluation-harness) to ensure fair comparisons. For API-based models, we used exact match evaluations of the generated outputs.
-The evaluation scores of ALLaM can be found in JSON format [here](https://huggingface.co/ALLaM-AI/ALLaM-7B-Instruct-preview/tree/main/evaluation).
 | Model                       |AVG | ETEC <br>0 shot   | IEN-MCQ <br>0 shot   | IEN-TF <br>0 shot   | AraPro <br>0 shot   | AraMath <br>5 shot   | Ar-IFEval <br>(prompt strict) <br>0 shot   | Ar-IFEval <br>(inst strict) <br>0 shot   | ExamsAR <br>5 shot   | ACVA  <br> 5 shot   | Arabic MMLU <br>0 Shot   | Openai MMLU <br>0 shot   | GAT  <br>0 shot   |

 All models were evaluated using our proprietary evaluation pipeline and [LM Evaluation Harness framework](https://github.com/EleutherAI/lm-evaluation-harness) to ensure fair comparisons. For API-based models, we used exact match evaluations of the generated outputs.
+The evaluation scores of ALLaM can be found in JSON format [here](https://huggingface.co/ALLaM-AI/ALLaM-7B-Instruct-preview/tree/main/evaluations).
 | Model                       |AVG | ETEC <br>0 shot   | IEN-MCQ <br>0 shot   | IEN-TF <br>0 shot   | AraPro <br>0 shot   | AraMath <br>5 shot   | Ar-IFEval <br>(prompt strict) <br>0 shot   | Ar-IFEval <br>(inst strict) <br>0 shot   | ExamsAR <br>5 shot   | ACVA  <br> 5 shot   | Arabic MMLU <br>0 Shot   | Openai MMLU <br>0 shot   | GAT  <br>0 shot   |