Instructions to use nvidia/Riva-Translate-4B-Instruct-v2 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use nvidia/Riva-Translate-4B-Instruct-v2 with Transformers:

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

pipe = pipeline("text-generation", model="nvidia/Riva-Translate-4B-Instruct-v2")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("nvidia/Riva-Translate-4B-Instruct-v2")
model = AutoModelForCausalLM.from_pretrained("nvidia/Riva-Translate-4B-Instruct-v2", 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=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use nvidia/Riva-Translate-4B-Instruct-v2 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "nvidia/Riva-Translate-4B-Instruct-v2"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "nvidia/Riva-Translate-4B-Instruct-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/nvidia/Riva-Translate-4B-Instruct-v2

SGLang

How to use nvidia/Riva-Translate-4B-Instruct-v2 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 "nvidia/Riva-Translate-4B-Instruct-v2" \
    --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": "nvidia/Riva-Translate-4B-Instruct-v2",
		"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 "nvidia/Riva-Translate-4B-Instruct-v2" \
        --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": "nvidia/Riva-Translate-4B-Instruct-v2",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use nvidia/Riva-Translate-4B-Instruct-v2 with Docker Model Runner:
```
docker model run hf.co/nvidia/Riva-Translate-4B-Instruct-v2
```

Riva-Translate-4B-Instruct-v2

File size: 2,363 Bytes

c0fe7b0

Field                                                                                                  |  Response
:------------------------------------------------------------------------------------------------------|:---------------------------------------------------------------------------------
Intended Task/Domain:                                                                                  |  Text-to-Text translation
Model Type:                                                                                            |  Decoder-only Transformer
Intended Users:                                                                                        |  Translators, marketers, and web developers who deliver content in multiple languages.
Output:                                                                                                |  Text translated into the target language
Describe how the model works:                                                                          |  The model takes text in the source language as input and translates it into the target language as text.
Name the adversely impacted groups this has been tested to deliver comparable outcomes regardless of:  |  Not Applicable
Technical Limitations & Mitigation:                                                                    |  Accuracy varies based on the characteristics of the input (domain, use case, noise, context, etc.). Grammatical errors and semantic issues may be present. As a potential mitigation, users can adjust the prompt to improve translation quality.
Verified to have met prescribed NVIDIA quality standards:  |  Yes
Performance Metrics:                                                                                   |  BLEU and COMET scores.
Potential Known Risks:                                                                                 |  Translations may not be 100% accurate. This is not recommended for word-for-word translation.
Licensing:                                                                                             |  GOVERNING TERMS: Use of the model is governed by the [NVIDIA Open Model License Agreement](https://www.nvidia.com/en-us/agreements/enterprise-software/nvidia-open-model-license/) ADDITIONAL INFORMATION: [Apache License, Version 2.0](https://www.apache.org/licenses/LICENSE-2.0).