Instructions to use beyoru/Opera with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use beyoru/Opera with Transformers:

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

pipe = pipeline("text-generation", model="beyoru/Opera")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
pipe(text=messages)

# Load model directly
from transformers import AutoProcessor, AutoModelForMultimodalLM

processor = AutoProcessor.from_pretrained("beyoru/Opera")
model = AutoModelForMultimodalLM.from_pretrained("beyoru/Opera", device_map="auto")
messages = [
    {
        "role": "user",
        "content": [
            {"type": "image", "url": "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/p-blog/candy.JPG"},
            {"type": "text", "text": "What animal is on the candy?"}
        ]
    },
]
inputs = processor.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(processor.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use beyoru/Opera with vLLM:

Install from pip and serve model

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

Use Docker

docker model run hf.co/beyoru/Opera

SGLang

How to use beyoru/Opera 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 "beyoru/Opera" \
    --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": "beyoru/Opera",
		"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 "beyoru/Opera" \
        --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": "beyoru/Opera",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use beyoru/Opera with Docker Model Runner:
```
docker model run hf.co/beyoru/Opera
```

beyoru commited on 17 days ago

Commit

7b6bceb

verified ·

1 Parent(s): 8e89934

Update README.md

Browse files

Files changed (1) hide show

README.md +0 -3

README.md CHANGED Viewed

@@ -20,9 +20,6 @@ and a composite reward: rule-based (function name + argument-subset partial cred
 structural tool-call validity, and **Intuitor self-certainty** (arXiv:2505.19590),
 rule and self-certainty weighted 1.0 each (normalized separately, then combined).
-- Base: Qwen/Qwen3.6-27B · Method: GRPO + LoRA (r32, alpha64, all-linear), merged to full weights
-- Tool-call format: qwen3_coder (`<tool_call><function=name><parameter=k>v</parameter></function></tool_call>`)
-- Reasoning: enabled (thinking ON)
 ## Serve (vLLM >= 0.25, no --trust-remote-code)
 ```bash

 structural tool-call validity, and **Intuitor self-certainty** (arXiv:2505.19590),
 rule and self-certainty weighted 1.0 each (normalized separately, then combined).
 ## Serve (vLLM >= 0.25, no --trust-remote-code)
 ```bash