Designed, trained and optimized on high-quality multilingual data, TowerLLM delivers best-in-class performance in translation and a wide range of translation tasks.
Ann Babel, AI-powered TowerLLM is a language operations (LangOps) platform that helps companies deliver multilingual customer experiences at scale. Large-scale language models (LLM) is designed, trained and optimized for translation, making it the best performing translation LLM on the market. Unbabel customers who use TowerLLM benefit from significantly improved machine translation accuracy, reduced errors and costs, and more cost-effective pricing than generic LLMs.
Top AI ML News: Honeywell Integrates AI into Guided Work Solutions to Improve Retail Performance
“Despite some doubts from some in the industry, TowerLLM clearly demonstrates that LLM is the best solution for machine translation.”
Developed from the ground up with multilingual support in mind, TowerLLM delivers better translation quality than leading LLMs such as GPT-4o and GPT-3.5, and consistently outperforms Google and DeepL. Built on large public datasets and trained using only the highest quality translation data filtered and curated by COMETKiwi, Unbabel's high-quality LLM, TowerLLM marks the next step in the evolution of AI translation.
“Despite skepticism from some in the industry, TowerLLM clearly demonstrates that LLM is the best solution for machine translation,” said João Graça, co-founder and chief technology officer at Unbabel. “We've invested years in developing LLM, so it's no surprise that TowerLLM outperformed GPT-4o. This is our second multilingual LLM we've released after COMETKiwi, and we're just getting started.”
read more: Data Observability + DataOps + FinOps: Towards a Healthy Data Adoption
During the release period, TowerLLM will provide:
- Machine translation across 18 language pairs
- Named entity recognition for localizing names, metrics and codes (currency, weight, location, brand, etc.)
- Source correction to eliminate grammatical and spelling errors in pre-translation
- Machine post-editing to automatically improve machine translations based on AI quality assessments that spotlight specific errors
Unbabel's experiments show that TowerLLM outperforms GPT-4o, GPT-3.5, Google, and DeepL in translation between language pairs and complex domains such as medicine, finance, technology, and law, and outperforms other LLMs in reasoning and contextual understanding. For more information and to test TowerLLM, please visit our dedicated landing page.
Unbabel is a pioneer in translation LLM and plans to release more language pairs and features in the coming months that will perform tasks that streamline translation workflows and improve quality. TowerLLM is available to customers only.
read more: Amazon unveils new 'private investigator' AI model
[To share your insights with us as part of editorial or sponsored content, please write to psen@itechseries.com]
