2+ day, 22+ hour ago  (28+ words) Mistral AI An updated Lean 4 formal proof engineering model optimised for automated theorem proving and autoformalization. 119 B total parameters, 6. 5 B active. Mistral AI " 2026...

3+ day, 9+ hour ago  (79+ words) All notable changes to Search Toolkit are documented here. Search Toolkit now uses a unified document model built around Document and Document Chunk, with a deterministic identity derived from a source_id and a locator. See the Document model page for full…...

1+ mon, 6+ day ago  (66+ words) Manage and Deploy Vespa Applications Mistral AI Build and operate Vespa applications with the Search Toolkit plugin. Define schemas via Python migrations, configure ranking and query profiles, deploy with the mistral-vespa CLI, and manage production clusters. Follow these guides to…...

1+ mon, 6+ day ago  (127+ words) Storage backends persist processed chunks and enable efficient search across your document collection. Vector stores enable semantic search by storing chunk embeddings and finding similar vectors. Use Vespa as a vector store in Search Toolkit ingestion and retrieval pipelines. Vespa…...

1+ mon, 6+ day ago  (334+ words) This page explains the concepts that make up a Vespa application: the application package, schemas, fields, ranking profiles, and query profiles. To learn how to create and configure these, see Manage Schema. A Vespa application package is a set of…...

1+ mon, 6+ day ago  (174+ words) Manage your Vespa application schemas through Python migrations. For the concepts behind schemas, fields, and ranking, see Anatomy of a Vespa application. Use migrations to define and modify your schemas. Migrations are Python files under vespa_app/migrations/ that are append-only, ordered…...

1+ mon, 6+ day ago  (226+ words) On-behalf-of workflows require a hardened deployment. They cannot be registered in a non-hardened deployment. Some workflows need to act as the user who triggered them " for example, accessing the user's connectors, querying resources scoped to their workspace, or calling the…...

1+ mon, 6+ day ago  (490+ words) Files bring source material into a task. Canvas is the built-in editor where Work produces, previews, and refines outputs. Together, they let you go from a raw upload (CSV, PDF, screenshot, document) to a polished, editable deliverable without leaving the…...

1+ mon, 6+ day ago  (180+ words) Skills are reusable instruction sets that extend Vibe Code with new workflows, custom slash commands, and scoped tool sets. Vibe follows the Agent Skills specification, so skills you write here are portable to other agents that adopt the same spec....

1+ mon, 6+ day ago  (180+ words) LLMs are not trained on your private data. To ground their answers in your documents you need a retrieval pipeline that ingests, indexes, and searches your content. Search Toolkit gives you every building block to assemble that pipeline. Ingestion transforms…...