Instructions to use joseph-ai/Aris-140M with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use joseph-ai/Aris-140M with Transformers:

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

pipe = pipeline("text-generation", model="joseph-ai/Aris-140M")
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("joseph-ai/Aris-140M")
model = AutoModelForCausalLM.from_pretrained("joseph-ai/Aris-140M", 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 joseph-ai/Aris-140M with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "joseph-ai/Aris-140M"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "joseph-ai/Aris-140M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/joseph-ai/Aris-140M

SGLang

How to use joseph-ai/Aris-140M 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 "joseph-ai/Aris-140M" \
    --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": "joseph-ai/Aris-140M",
		"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 "joseph-ai/Aris-140M" \
        --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": "joseph-ai/Aris-140M",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use joseph-ai/Aris-140M with Docker Model Runner:
```
docker model run hf.co/joseph-ai/Aris-140M
```

Aris-140M

This model is a fine-tuned version of on an unknown dataset. It achieves the following results on the evaluation set:

Loss: 3.3110

Model description

More information needed

Intended uses & limitations

More information needed

Training and evaluation data

More information needed

Training procedure

Training hyperparameters

The following hyperparameters were used during training:

learning_rate: 0.00025
train_batch_size: 2
eval_batch_size: 2
seed: 42
distributed_type: multi-GPU
num_devices: 2
gradient_accumulation_steps: 16
total_train_batch_size: 64
total_eval_batch_size: 4
optimizer: Use OptimizerNames.ADAMW_TORCH with betas=(0.9,0.95) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
lr_scheduler_type: cosine
lr_scheduler_warmup_steps: 200
training_steps: 7630

Training results

Training Loss	Epoch	Step	Validation Loss
5.0026	0.0655	500	4.9335
4.1512	0.1311	1000	4.1450
3.9029	0.1966	1500	3.8596
3.7558	0.2621	2000	3.7075
3.5920	0.3277	2500	3.6100
3.5657	0.3932	3000	3.5328
3.4994	0.4587	3500	3.4764
3.4811	0.5242	4000	3.4332
3.4043	0.5898	4500	3.3970
3.3963	0.6553	5000	3.3697
3.3784	0.7208	5500	3.3471
3.3817	0.7864	6000	3.3304
3.3694	0.8519	6500	3.3189
3.3337	0.9174	7000	3.3129
3.3679	0.9830	7500	3.3111
3.3503	1.0	7630	3.3110

Framework versions

Transformers 5.12.1
Pytorch 2.6.0+cu124
Datasets 5.0.0
Tokenizers 0.22.2

Downloads last month: 7

Safetensors

Model size

0.1B params

Tensor type

F32