Task-Specific Models overview
Dedicated APIs that work out-of-the-box for specific reading and writing tasks.
These models are no longer supported. For better results, see Jamba 1.5.
AI21 Studio's Task-Specific Models (TSMs) offer a range of powerful tools. These models have been specifically designed for their respective tasks and provide high-quality results while optimizing efficiency.
Here is a list of our Task-Specific Models:
| Model | Description |
|---|---|
| Contextual Answers (In-prompt) | Provide information and a question in the prompt, will return an answer based solely on the prompt information. If the answer to your question is not in the prompt, the model will not attempt to provide an answer. |
| Contextual answers (RAG library) | Ask a question based on specified documents uploaded to your document library. The model will answer your question based solely on information in the specified documents. If the answer to your question is not in the documents, the model will not attempt to provide an answer. |
| Semantic Search | Searches your RAG Engine library for documents that have information about the provided topics or keywords. |
| Embeddings | Generates a vector representation of the provided text. |
| Paraphrase | Suggests alternative ways to rewrite a given message using different words, in a suggested tone and style. |
| Grammatical Error Corrections | Identifies and corrects grammar errors, returning a list of grammar errors and suggestions for corrections. |
| Text Improvements | Scans a piece of text to see if it can be improved, checking for fluency, clarity, and vocabulary. |
| Summarize | Given text or a URL, generates a summary based on the provided text or fetched page. |
| Summarize by Segment | Splits text into topical segments and provides a summary of each segment. |
| Text Segmentation | Breaks down a piece of text into segments, identifying distinct topics and lines that will work well together and form a coherent piece of text. |
Updated 10 days ago