In the last month, Sensible upgraded its document-type classification endpoints to use large language model (LLM) completions instead of embeddings-based scorings, updated its GTP-4 vision model for better multimodal data extraction results, enabled uploading an unlimited number of reference documents to the web app, and added advanced functionality to several layout-based and computed field methods.

In the last month Sensible added support for extracting and classifying CSV files, made improvements to the LLM-based List method, added advanced support for extended JsonLogic operations, and added several advanced configuration options.

In the last month, we released a powerful new feature for configuring extraction from multi-document files ("portfolios") using large language models (LLMs). We also released the capability to specify a custom schema for your extracted document data instead of Sensible's default parsed_document schema. We improved the Sensible app user experience by enabling search for field keys and values in a new, compact view of extracted data. We also updated several model versions for LLM-based methods.

In the last, month we made both an improvement and a breaking change to our support for extracting from Excel documents. We also updated our large language model (LLM) support for the List method and Query Group method in response to end-of-life GPT versions.

In the last month, we released a new feature for visualizing extracted data in the JSON editor and advanced configuration for several layout-based methods.

In the last month, we released a major new feature to enable manual review and correction to your document extraction process. We also released support for an Anthropic large language model (LLM) and support for a cutting-edge LLM approach to locating the answer to your prompts in the document.

In the last month, we made UX improvements to the Sensible app, added advanced configurability to the Intersection method, and added a new preprocessor.

In the last month, we made several improvements to the web app user experience and deprecated the Topic method.

In the last month we released support for GPT-4 Vision, so you can extract information from non-text images in documents. We also released a new recommend query groups feature, so you can upload and extract data from a document with a few button clicks instead of having to author your own LLM prompts. We added a new filterable extraction status, deprecated several methods, and made several improvements to advanced configurability.

In the last month, we released support for extracting data from Excel files and added Google sign-in support to the Sensible app. To teach users how to author LLM-powered queries, we added example extractions to the app. We deprecated the Query method and replaced it with the Query Group method to improve LLM performance, deprecated the Table method and replaced it with the NLP Table method, and made several minor improvements.