Instructions to use optimum-intel-internal-testing/tiny-random-ouro with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use optimum-intel-internal-testing/tiny-random-ouro with Transformers:

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

pipe = pipeline("text-generation", model="optimum-intel-internal-testing/tiny-random-ouro", trust_remote_code=True)
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoModelForCausalLM
model = AutoModelForCausalLM.from_pretrained("optimum-intel-internal-testing/tiny-random-ouro", trust_remote_code=True, dtype="auto")

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use optimum-intel-internal-testing/tiny-random-ouro with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/optimum-intel-internal-testing/tiny-random-ouro

SGLang

How to use optimum-intel-internal-testing/tiny-random-ouro 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 "optimum-intel-internal-testing/tiny-random-ouro" \
    --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": "optimum-intel-internal-testing/tiny-random-ouro",
		"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 "optimum-intel-internal-testing/tiny-random-ouro" \
        --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": "optimum-intel-internal-testing/tiny-random-ouro",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use optimum-intel-internal-testing/tiny-random-ouro with Docker Model Runner:
```
docker model run hf.co/optimum-intel-internal-testing/tiny-random-ouro
```

tiny-random-ouro / README.md

snake7gun

Upload folder using huggingface_hub

0bbc6df verified 3 days ago

preview code

raw

history blame contribute delete

768 Bytes

	---
	library_name: transformers
	license: apache-2.0
	---

	# tiny-random-ouro

	Tiny random [Ouro](https://huggingface.co/ByteDance/Ouro-1.4B) model for `optimum-intel` OpenVINO CI.

	Architecture: `OuroForCausalLM` (Universal Transformer). Generated with random weights; the full
	`vocab_size` (49152) and the real tokenizer are kept so the tokenizer round-trips, while all other
	dimensions are shrunk. `total_ut_steps=4` is preserved to exercise the Universal-Transformer loop.

	\| field \| value \|
	\|-------\|-------\|
	\| hidden_size \| 32 \|
	\| intermediate_size \| 64 \|
	\| num_hidden_layers \| 2 \|
	\| num_attention_heads \| 4 \|
	\| num_key_value_heads \| 2 \|
	\| head_dim \| 8 \|
	\| total_ut_steps \| 4 \|
	\| vocab_size \| 49152 \|

	Reproduce with the script in the optimum-intel PR description.