Text retrieval, Semantic similarity, Clustering
Amazon Titan Embeddings G1 – Text v1.2 is an advanced embedding model designed to excel in various text-based tasks with high performance and precision. Capable of processing up to 8,000 tokens and generating vectors of 1,536 dimensions, this model is tailored for text retrieval applications, offering enhanced capabilities for semantic similarity and clustering. Its robust architecture allows it to effectively handle long documents, though it is advisable to segment documents into logical parts, such as paragraphs or sections, for optimal retrieval performance.
Supporting over 25 languages, Titan Embeddings G1 – Text v1.2 is well-suited for diverse linguistic applications and global data analysis. The model’s high dimensional vector output and ability to manage extensive token inputs make it a versatile tool for complex text processing needs. It is particularly useful for organizations and developers seeking advanced solutions for text retrieval, semantic analysis, and clustering tasks across a range of languages and document types.
Amazon Titan Embeddings G1 – Text v1.2 represents a powerful choice for high-performance text embedding needs, offering comprehensive capabilities and extensive language support for advanced text-based applications.
Below you will find all supported platforms and the related CogniTech AI Credits costs.
| Region | Context | TPM | RPM | Input Credits | Output Credits |
|---|---|---|---|---|---|
| Region | Context | TPM | RPM | Input Credits | Output Credits |
| us-west-2 | 8,192 | 300,000 | 2,000 | 0.013 / 1000 tokens | 0 / 1000 tokens |