Riva-Translate-4B-Instruct-v2 NIM Overview

Description:

The Riva-Translate-4B-Instruct-v2 Neural Machine Translation NIM translates text in English and 36 non-English languages. The supported languages are: English(en), Czech(cs), Danish(da), German(de), Greek(el), European Spanish(es-ES), LATAM Spanish(es-US), Finnish(fi), French(fr), Hungarian(hu), Italian(it), Lithuanian(lt), Latvian(lv), Dutch(nl), Norwegian(no), Polish(pl), European Portuguese(pt-PT), Brazilian Portuguese(pt-BR), Romanian(ro), Russian(ru), Slovak(sk), Swedish(sv), Simplified Chinese(zh-CN), Traditional Chinese(zh-TW), Japanese(ja), Hindi(hi), Korean(ko), Estonian(et), Slovenian(sl), Bulgarian(bg), Ukrainian(uk), Croatian(hr), Arabic(ar), Vietnamese(vi), Turkish(tr), Indonesian(id), Thai(th). It supports both sentence- and document-level translation.

This container houses a single model program class: the Riva-Translate-4B-Instruct-v2 neural machine translation model, embedded and served for GPU-accelerated inference.

Translators, marketers, and web developers who deliver content in multiple languages are the primary audience for this container.

The container components are ready for commercial or non-commercial use.

License/Terms of Use:

GOVERNING TERMS: Use of this container is governed by the NVIDIA Software License Agreement and Product-Specific Terms for NVIDIA AI Products.

The use of any models in this container is governed by the NVIDIA Open Model License Agreement.

You are responsible for ensuring that your use of any provided models complies with all applicable laws.

ADDITIONAL INFORMATION: Apache License, Version 2.0.

Deployment Geography:

Global

Release Date:

Huggingface 07/17/2026 via nvidia/riva-translate-4b-instruct-v2

Program Classes:

The Riva-Translate-4B-Instruct-v2 NIM Container includes the following model:

Deployment Details:

This downloadable NIM packages the Riva-Translate-4B-Instruct-v2 model for GPU-accelerated inference via an OpenAI-compatible HTTP API on port 8000. The container supports amd64 and arm64 architectures. The model is embedded in the container; no separate model download step is required at runtime.

Our AI models are designed and/or optimized to run on NVIDIA GPU-accelerated systems. By leveraging NVIDIA’s hardware (e.g. GPU cores) and software frameworks (e.g., CUDA libraries), the model achieves faster training and inference times compared to CPU-only solutions.

Prerequisites

Before deploying, ensure your host has a Linux operating system, an NVIDIA GPU, Docker 24 or later, the NVIDIA Container Toolkit, and an NGC Personal API Key. Accept the governing terms on the NGC container page for this NIM before your first pull. For prerequisites, installation, and full deployment steps, refer to the NVIDIA NIM for Large Language Models documentation.

Pull and run

Set CONTAINER_IMAGE to the pull URI shown on the NGC container page for this NIM, then authenticate with NGC, pull the image, prepare a local model cache, and start the NIM:

# Required: set this to the NGC pull URI from the container page.
# Example form: nvcr.io/<org>/<team>/riva-translate-4b-instruct-v2:<version>
export CONTAINER_IMAGE="<NGC_CONTAINER_IMAGE>"

echo "$NGC_API_KEY" | docker login nvcr.io --username '$oauthtoken' --password-stdin

docker pull "$CONTAINER_IMAGE"

export LOCAL_NIM_CACHE=~/.cache/nim
mkdir -p "$LOCAL_NIM_CACHE"

docker run --gpus=all \
  -e NGC_API_KEY="$NGC_API_KEY" \
  -v "$LOCAL_NIM_CACHE:/opt/nim/.cache" \
  --shm-size=16g \
  -p 8000:8000 \
  "$CONTAINER_IMAGE"

Verify readiness

Confirm the server is live and the model is loaded:

curl -s http://localhost:8000/v1/health/ready
curl -s http://localhost:8000/v1/models

The id field in the /v1/models response is the model name to use in API requests.

Invoke translation

Specify the source and target language pair in the system message using tags such as en-de (English to German), en-zh-cn (English to Simplified Chinese), de-en (German to English), fr-en (French to English), or zh-cn-en (Simplified Chinese to English). See the model card for the full list of supported language-pair tags.

Send a chat completion request:

curl http://localhost:8000/v1/chat/completions \
  -H "Content-Type: application/json" \
  -d '{
    "model": "nvidia/riva-translate-4b-instruct-v2",
    "messages": [
      {"role": "system", "content": "en-zh-cn"},
      {"role": "user", "content": "The GRACE mission is a collaboration between NASA and the German Aerospace Center."}
    ],
    "max_tokens": 256
  }'

We recommend using the system-message language-pair tag format shown above; the model may not perform optimally without it.

Operating System: Linux

Reference(s):

[1] Vaswani, Ashish, et al. "Attention is all you need." arXiv preprint arXiv:1706.03762 (2017).
[2] https://github.com/openai/tiktoken
[3] https://en.wikipedia.org/wiki/BLEU
[4] https://github.com/mjpost/sacreBLEU
[5] https://github.com/Unbabel/COMET
[6] NVIDIA NeMo Toolkit
[7] Riva-Translate-4B-Instruct-v2 Model Card — model architecture, training dataset disclosures, performance benchmarks, supported hardware, and inference details

Security Common Vulnerabilities and Exposures (CVEs)

Please review the Security Scanning tab on NGC to view the latest security scan results. For certain open-source vulnerabilities listed in the scan results, NVIDIA provides a response in the form of a Vulnerability Exploitability eXchange (VEX) document. The VEX information can be reviewed and downloaded from the Security Scanning tab.

Ethical Considerations:

NVIDIA believes Trustworthy AI is a shared responsibility and we have established policies and practices to enable development for a wide array of AI applications. Developers should work with their internal developer team to ensure these software components meet requirements for the relevant industry and use case and address unforeseen product misuse.

Please report quality, risk, security vulnerabilities or NVIDIA AI Concerns here.

Get Help

Getting started with the NIM

Deploying and integrating the NIM is straightforward thanks to our industry standard APIs. For release documentation, prerequisites, installation, quickstart, and advanced deployment guides, refer to the NVIDIA NIM for Large Language Models documentation, starting with the Quickstart guide on that page. After the container is published, also use the NGC container page for this NIM for the exact pull URI and release notes.

NVIDIA Developer Community Forum

Get access to community knowledge base articles and support cases (https://forums.developer.nvidia.com/)

Enterprise Support

Get access to knowledge base articles and support cases or submit a ticket.