Feed

Latest Intelligence

Every AI development we have covered, newest first. Filter by section to focus on what matters to you.

Hugging FaceJun 30, 2026Enterprise

ScarfBench: Benchmarking AI Agents for Enterprise Java Framework Migration

IBM Research has introduced ScarfBench, a specialised benchmark designed to evaluate autonomous AI agents on enterprise Java framework migration tasks, published via Hugging Face. The benchmark tests the capability of AI models and agentic workflows to refactor, upgrade, and modernise legacy enterprise codebases across complex Java application frameworks. It establishes standardized criteria for measuring agent accuracy, code consistency, and autonomous migration performance in legacy enterprise environments.

22/100

Microsoft ResearchJun 30, 2026Models

SkillOpt: Agent skills as trainable parameters

Microsoft Research introduced SkillOpt, a framework that treats AI agent instructions and skills as trainable parameters rather than manually adjusted prompts. The technique formalizes skill refinement into an automated optimization process, enabling agents to systematically improve task performance and reliability without modifying underlying base model weights.

25/100

Google DeepMindJun 30, 2026Models

Start building with Nano Banana 2 Lite and Gemini Omni Flash

Google DeepMind announced the availability of two new models for developers: Nano Banana 2 Lite and Gemini Omni Flash. According to the announcement, the releases target developers seeking lightweight and multimodal foundation models for application development. Full technical specifications and benchmark data were not detailed in the metadata, but the release signals an expansion of Google's accessible model tiers for on-device and low-latency cloud inference use cases.

34/100

Hugging FaceJun 30, 2026Models

Featuring Every Eval Ever Results on Hugging Face Model Pages

Hugging Face has introduced direct integration of benchmark results from the Every Eval Ever initiative onto its model repository pages. This update provides standardized evaluation metrics directly within individual model listings, allowing users to assess and compare performance across various benchmarks without navigating external testing platforms. The feature aims to streamline discovery and due diligence for open-source and hosted machine learning models across the platform.

31/100

Google DeepMindJun 24, 2026Models

Introducing computer use in Gemini 3.5 Flash

Google DeepMind announced computer use capabilities for its Gemini 3.5 Flash model, enabling the lightweight AI system to interact directly with graphical user interfaces, navigate operating systems, and execute multi-step workflows. According to the company, bringing computer control to the Flash tier expands automated interface interaction to a lower-latency, more cost-effective model architecture compared to flagship foundation models.

73/100

Hugging FaceJun 24, 2026Enterprise

Accelerating Transformers Fine-Tuning with NVIDIA NeMo AutoModel

Nvidia and Hugging Face have published a technical integration detailing how to accelerate Transformer model fine-tuning using Nvidia NeMo AutoModel. The collaboration focuses on streamlining the adaptation of large language models by integrating NeMo's distributed training and optimization capabilities directly within Hugging Face workflows. The integration is designed to reduce training times and compute overhead for developers working with open model architectures across standard enterprise and research environments.

31/100

The RegisterJun 23, 2026Enterprise

Anthropic reimagines Claude in Slack as nosy, always-on agentic AI coworker

Anthropic has revamped its Claude integration for Slack, shifting the application from a reactive chat assistant into an autonomous, always-on agentic coworker. The updated integration is designed to participate more actively within Slack workspaces, monitoring conversation context and executing tasks proactively. This shift reflects a broader strategy by Anthropic to embed persistent AI agents directly into everyday enterprise workflow and collaboration environments.

40/100

Mistral AIJun 23, 2026Enterprise

Introducing Mistral OCR 4

Mistral AI has announced the launch of Mistral OCR 4, an enterprise-focused document artificial intelligence model. According to the vendor, the release features optical character recognition across 170 languages, bounding box detection for layout parsing, and support for self-hosted deployment options. The model is designed to process complex document structures, enabling organizations to extract structured information from unstructured files within secure, private infrastructure environments.

38/100

Hugging FaceJun 23, 2026Enterprise

Shipping huggingface_hub every week with AI, open tools, and a human in the loop

Hugging Face detailed its engineering workflow for maintaining and releasing the huggingface_hub library on a weekly cadence. The process integrates automated AI-assisted tooling, open-source utilities, and human-in-the-loop oversight to manage continuous integration and deployment. According to the company, combining automated generation tools with manual verification allows the team to accelerate release cycles, handle routine dependency updates, and maintain software reliability across the primary client library for the Hugging Face platform.

27/100

Hugging FaceJun 22, 2026Models

PP-OCRv6 on Hugging Face: 50-Language OCR from 1.5M to 34.5M Parameters

PaddlePaddle has released PP-OCRv6 on Hugging Face, introducing a suite of open-weight optical character recognition models spanning 1.5M to 34.5M parameters. The new iteration provides multi-language OCR capabilities across 50 languages. Designed for lightweight edge and high-throughput document processing pipelines, the model family offers varying parameter tiers to balance latency, memory constraints, and transcription accuracy across diverse deployment environments.

25/100