
Nous Research has added a Blank Slate setup mode to its open-source Hermes Agent. It inverts the usual onboarding. Instead of a fully loaded default, you start with almost nothing. Hermes Agent is the self-improving agent framework from Nous Research. It runs on your own machine. The team announced the new mode on X. Blank Slate now joins two existing options: Quick Setup and Full Setup. TL;DR Blank Slate boots an agent with everything off except provider & model, File Operations,
Will Nous Research's Hermes Agent repo reach 500 GitHub stars by June 28, 2026?
Resolves by Jun 28, 2026
Nous Research has added a Blank Slate setup mode to its open-source Hermes Agent framework, which starts with minimal features enabled rather than a fully loaded default. The mode enables only a provider, model, file operations, and terminal access, while keeping web, browser, code execution, vision, memory, and other advanced tools disabled by default. This approach matters because it writes configuration decisions to disk in a way that persists across updates, ensuring that disabled features will not silently re-enable later. The Blank Slate mode serves security-sensitive deployments, reproducible team setups, and teaching environments where users want explicit control over which capabilities their agent can access.

Today, Mistral AI released OCR 4, its latest document-understanding model. This new release adds bounding boxes, block classification, and inline confidence scores alongside extracted text. It supports 170 languages across 10 language groups and runs in a single container for fully self-hosted deployments. OCR 4 also serves as an ingestion component for enterprise search, RAG, and domain-specific retrieval pipelines. TL;DR OCR 4 returns bounding boxes, typed-block labels, and per-word c

Datalab has released lift, a 9B open-weights vision model for structured extraction. You pass it a JSON schema, and it returns a JSON object that matches. The model reads PDFs and images directly, then decodes against your schema. This is Datalab’s first model built purely for extraction. The team already ships open-source OCR tools: chandra, marker, and surya. lift extends that work into schema-driven field extraction. lift scores 90.2% field accuracy on Datalab’s 225-documen

Mistral OCR 4 delivers enterprise document AI with 170-language support, bounding boxes, and self-hosted deployment.
Want to go deeper than the news? Explore live, cohort-based AI courses taught by practitioners.
Browse AI courses on Maven