ynklab/Tower-7B-d2d

TEXT GENERATIONPricing:Input $0.4 / Cached $0.08 / Output $0.8Concurrent Unit Cost:1Model Size:7BQuant:FP8Context Size:4kTool Calling:SupportedPublished:Sep 7, 2026License:cc-by-nc-sa-4.0Architecture:Transformer Open Weights Featherless Exclusive Cold

ynklab/Tower-7B-d2d is a 7 billion parameter language model developed by ynklab, fine-tuned from Unbabel/TowerInstruct-Mistral-7B-v0.2. This model is specifically optimized for multilingual document-level machine translation, trained directly on document-to-document (d2d) translation tasks using the DocBlocks dataset. It supports translation between English and nine other languages, making it suitable for applications requiring coherent, long-form text translation.

Loading preview...

Tower-7B-d2d: Document-Level Machine Translation Model

Tower-7B-d2d is a 7 billion parameter language model developed by ynklab, specifically fine-tuned for multilingual document-level machine translation. It is based on the Unbabel/TowerInstruct-Mistral-7B-v0.2 architecture and was trained using full-parameter supervised fine-tuning on the DocBlocks dataset.

Key Capabilities

  • Document-to-Document (d2d) Translation: Optimized for translating entire documents, ensuring length consistency and contextual coherence across longer texts.
  • Multilingual Support: Facilitates translation between English and nine other languages: German, Spanish, French, Italian, Korean, Dutch, Portuguese, Russian, and Chinese, in both directions.
  • High Context Length: Supports a maximum sequence length of 32,768 tokens, allowing for the processing of substantial document segments.
  • ChatML Prompt Format: Designed to be used with a specific ChatML-style prompt for translation tasks, ensuring optimal performance.

When to Use This Model

  • Document Translation: Ideal for applications requiring the translation of full documents or long texts where maintaining context and consistency across sentences is crucial.
  • Multilingual Applications: Suitable for systems needing robust translation capabilities between English and the supported languages.
  • Research in MT: Can be used as a strong baseline or component in research related to document-level machine translation, as detailed in the associated Doc2FRC paper and GitHub repository.

Finetunes
1 models