Sold by: Widn.AI
Deployed on AWS
Free Trial
Tower Anthill by Widn is a powerful multilingual LLM designed for advanced translation tasks across multiple languages and domains.
Overview
Widn Tower Anthill is a multilingual LLM based on Unbabel's powerful Tower LLMs, optimized for high-quality translation use cases across multiple domains. It is the smallest and fastest model offered by Widn.
Highlights
- **Widn Tower Anthill** is intended for multilingual tasks and is specially strong on machine translation. This means that you should be able to solve several translation use cases that traditional NMT models struggle with. This includes: - Translation of entire documents; - Translation with few-shots for dynamic adaption; - Translation following specific terminologies/glossaries; - Translation into different tones; - Translation following style guides.
- **Widn Tower Anthill** was trained on a diverse multilingual dataset comprising millions of high-quality translations across various domains. While it excels in many languages, performance may vary for low-resource languages or highly specialized technical content.
- **Widn Tower Anthill** supports 26 languages and variants: German, Spanish, French, Italian, Korean, Dutch, Russian, English, Portuguese (Portugal), Portuguese (Brazilian), Spanish (Latin America), Chinese (Simplified), Chinese (Traditional), Czech, Ukrainian, Hindi, Icelandic, Japanese, Polish, Swedish, Hungarian, Romanian, Danish, Norwegian (Nynorsk), Norwegian (Bokmål), Finnish
Details
Sold by
Delivery method
Latest version
Deployed on AWS
New
Introducing multi-product solutions
You can now purchase comprehensive solutions tailored to use cases and industries.
Features and programs
Financing for AWS Marketplace purchases
AWS Marketplace now accepts line of credit payments through the PNC Vendor Finance program. This program is available to select AWS customers in the US, excluding NV, NC, ND, TN, & VT.
Pricing
Free trial
Try this product free for 7 days according to the free trial terms set by the vendor.
Pricing is based on actual usage, with charges varying according to how much you consume. Subscriptions have no end date and may be canceled any time.
Additional AWS infrastructure costs may apply. Use the AWS Pricing Calculator to estimate your infrastructure costs.
Dimension | Description | Cost/host/hour |
|---|---|---|
ml.g5.xlarge Inference (Batch) Recommended | Model inference on the ml.g5.xlarge instance type, batch mode | $10.00 |
ml.g5.xlarge Inference (Real-Time) Recommended | Model inference on the ml.g5.xlarge instance type, real-time mode | $10.00 |
Dimensions summary
You pay by the hour for model inference running on the ml.g5.xlarge instance type. This is usage-based billing, so charges reflect how many host hours you run. Two options separate the work by processing mode. Batch mode handles groups of translations processed together. Real-time mode handles translations returned as you request them. Both use the same instance size, so the mode you pick reflects your workload, not a capacity tier. You can run either or both depending on how you deliver translations.
Top-of-mind questions for buyers
You pay per host hour for the ml.g5.xlarge instance type. This is a GPU-backed instance that runs the translation model. Billing counts each hour the instance runs, in either batch or real-time mode. You are charged for running time, not for the number of translations processed.
Both meter the same instance per host hour. Batch mode processes grouped translations together, so you run the instance for the length of each batch job. Real-time mode returns translations as you request them, so the instance runs for the duration you keep it active. Your bill reflects host hours in either mode.
Charges apply per host hour while the instance runs. A fully stopped instance does not accrue these software host-hour charges. If you keep the instance running between jobs to stay ready, those hours still count. Underlying AWS storage or resource fees may apply separately even when stopped.
widn.ai
Helpful?
Vendor refund policy
No refunds. Please contact support+aws@widn.ai for further assistance.
Legal
Vendor terms and conditions
Upon subscribing to this product, you must acknowledge and agree to the terms and conditions outlined in the vendor's End User License Agreement (EULA) .
Content disclaimer
Vendors are responsible for their product descriptions and other product content. AWS does not warrant that vendors' product descriptions or other product content are accurate, complete, reliable, current, or error-free.
Delivery details
Amazon SageMaker model
An Amazon SageMaker model package is a pre-trained machine learning model ready to use without additional training. Use the model package to create a model on Amazon SageMaker for real-time inference or batch processing. Amazon SageMaker is a fully managed platform for building, training, and deploying machine learning models at scale.
Deploy the model on Amazon SageMaker AI using the following options:
Version release notes
This is an improved version of the first Widn Tower Anthill, a powerful multilingual LLM optimized for high-quality translation use cases across multiple domains, and now much better at following instructions.
Additional details
Inputs
- Summary
The model accepts JSON input containing a prompt with the text to be translated and optional model parameters.
See the notebook for examples of tested prompts.
- Limitations for input type
- The input text should be clear and well-formed. The maximum token limit is 4096 tokens. For best quality, use the prompt examples shown in the example and in the notebook.
- Input MIME type
- application/json
https://github.com/widn-ai/aws-marketplace/blob/main/data/input/real-time/example.json
https://github.com/widn-ai/aws-marketplace
Input data descriptions
The following table describes supported input data fields for real-time inference and batch transform.
Field name | Description | Constraints | Required |
|---|---|---|---|
messages.role | The role of the message. Examples: "system", "user", "assistant". | Type: FreeText
Limitations: Use "user" for better results. | Yes |
messages.content | The content of the message. Example: "Translate the following text from Portuguese into English.\n Portuguese: Um grupo de investigadores lançou um novo modelo para tarefas relacionadas com tradução.\n English:” | Type: FreeText
Limitations: Be aware of the max number of tokens supported (4096). | Yes |
Custom attributes
The following table describes custom attributes for real-time inference endpoints.
Field name | Description | Constraints | Required |
|---|---|---|---|
max_tokens | The maximum number of tokens that can be generated in the chat completion. This value can be used to control costs for text generated via API. | Default value: null
Type: Integer
Minimum: 1
Maximum: 4000 | No |
Resources
Vendor resources
Support
Vendor support
Contact us at support+aws@widn.ai .
AWS infrastructure support
AWS Support is a one-on-one, fast-response support channel that is staffed 24x7x365 with experienced and technical support engineers. The service helps customers of all sizes and technical abilities to successfully utilize the products and features provided by Amazon Web Services.
Similar products
Tower Vesuvius by Widn is a powerful multilingual LLM designed for advanced translation tasks across multiple languages and domains.
Tower Sugarloaf by Widn is a powerful multilingual LLM designed for advanced translation tasks across multiple languages and domains.
Widn's API is a RESTful API that lets you automate translation of text and documents, manage custom glossaries, estimate translation quality, and evaluate MT systems for a seamless, customizable AI-powered translation experience.
COMETKiwi provides state-of-the-art quality estimation for machine translations without requiring reference translations.
Customer reviews
No customer reviews yet
Be the first to review this product . We've partnered with PeerSpot to gather customer feedback. You can share your experience by writing or recording a review, or scheduling a call with a PeerSpot analyst.