News

MarkTechPost
marktechpost.com > 02/19/2026 > google-ai-releases-gemini-3-1-pro-with-1-million-token-context-and-77-1-percent-arc-agi-2-reasoning-for-ai-agents

Google AI Releases Gemini 3.1 Pro with 1 Million Token Context and 77.1 Percent ARC-AGI-2 Reasoning for AI Agents

1+ hour, 57+ min ago  (512+ words) Google has officially shifted the Gemini era into high gear with the release of Gemini 3.1 Pro, the first version update in the Gemini 3 series. This release is not just a minor patch; it is a targeted strike at the "agentic…...

MarkTechPost
marktechpost.com > 02/19/2026 > a-coding-implementation-to-build-bulletproof-agentic-workflows-with-pydanticai-using-strict-schemas-tool-injection-and-model-agnostic-execution

A Coding Implementation to Build Bulletproof Agentic Workflows with PydanticAI Using Strict Schemas, Tool Injection, and Model-Agnostic Execution

2+ hour, 57+ min ago  (262+ words) We set up the execution environment and ensure all required libraries are available for the agent to run correctly. We securely load the OpenAI API key in a Colab-friendly way so the tutorial works without manual configuration changes. We also…...

MarkTechPost
marktechpost.com > 02/18/2026 > zyphra-releases-zuna-a-380m-parameter-bci-foundation-model-for-eeg-data-advancing-noninvasive-thought-to-text-development

Zyphra Releases ZUNA: A 380M-Parameter BCI Foundation Model for EEG Data, Advancing Noninvasive Thought-to-Text Development

16+ hour, 20+ min ago  (291+ words) For decades, researchers have struggled with the "Wild West" of EEG data. Different datasets use varying numbers of channels and inconsistent electrode positions. Most deep learning models are trained on fixed channel montages, making them fail when applied to new…...

MarkTechPost
marktechpost.com > 02/18/2026 > tutorial-building-a-visual-document-retrieval-pipeline-with-colpali-and-late-interaction-scoring

[Tutorial] Building a Visual Document Retrieval Pipeline with ColPali and Late Interaction Scoring

22+ hour, 25+ min ago  (219+ words) We prepare a clean and stable execution environment by uninstalling conflicting packages and upgrading pip. We explicitly pin compatible versions of Pillow and torchaudio to avoid runtime import errors. We then install ColPali and its required dependencies so the rest…...

MarkTechPost
marktechpost.com > 02/18/2026 > tavus-launches-phoenix-4-a-gaussian-diffusion-model-bringing-real-time-emotional-intelligence-and-sub-600ms-latency-to-generative-video-ai

Tavus Launches Phoenix-4: A Gaussian-Diffusion Model Bringing Real-Time Emotional Intelligence And Sub-600ms Latency To Generative Video AI

23+ hour, 50+ min ago  (455+ words) The "uncanny valley" is the final frontier for generative video. We have seen AI avatars that can talk, but they often lack the soul of human interaction. They suffer from stiff movements and a lack of emotional context. Tavus aims…...

MarkTechPost
marktechpost.com > 02/18/2026 > google-deepmind-releases-lyria-3-an-advanced-music-generation-ai-model-that-turns-photos-and-text-into-custom-tracks-with-included-lyrics-and-vocals

Google DeepMind Releases Lyria 3: An Advanced Music Generation AI Model that Turns Photos and Text into Custom Tracks with Included Lyrics and Vocals

1+ day, 2+ hour ago  (515+ words) Google DeepMind is pushing the boundaries of generative AI again. This time, the focus is not on text or images. It is on music. The Google team recently introduced Lyria 3, their most advanced music generation model to date. Lyria 3 represents…...

MarkTechPost
marktechpost.com > 02/18/2026 > google-introduces-jetpack-compose-glimmer-a-new-spatial-ui-framework-designed-specifically-for-the-next-generation-of-ai-glasses

Google Introduces Jetpack Compose Glimmer: A New Spatial UI Framework Designed Specifically for the Next Generation of AI Glasses

1+ day, 14+ hour ago  (511+ words) Google is moving beyond the rectangular screen. For over 10 years, Google designers have explored how to build interfaces for transparent displays. The result is Jetpack Compose Glimmer, a design system built specifically for display AI glasses. For devs and data…...

MarkTechPost
marktechpost.com > 02/17/2026 > cohere-releases-tiny-aya-a-3b-parameter-small-language-model-that-supports-70-languages-and-runs-locally-even-on-a-phone

Cohere Releases Tiny Aya: A 3B-Parameter Small Language Model that Supports 70 Languages and Runs Locally Even on a Phone

1+ day, 16+ hour ago  (160+ words) Cohere AI Labs has released Tiny Aya, a family of small language models (SLMs) that redefines multilingual performance. While many models scale by increasing parameters, Tiny Aya uses a 3.35B-parameter architecture to deliver state-of-the-art translation and generation across 70 languages. Tiny…...

MarkTechPost
marktechpost.com > 02/17/2026 > anthropic-releases-claude-4-6-sonnet-with-1-million-token-context-to-solve-complex-coding-and-search-for-developers

Anthropic Releases Claude 4.6 Sonnet with 1 Million Token Context to Solve Complex Coding and Search for Developers

2+ day, 1+ hour ago  (441+ words) Anthropic is officially entering its "Thinking" era. Today, the company announced Claude 4.6 Sonnet, a model designed to transform how devs and data scientists handle complex logic. Alongside this release comes Improved Web Search with Dynamic Filtering, a feature that uses…...

MarkTechPost
marktechpost.com > 02/17/2026 > how-to-build-an-advanced-interactive-exploratory-data-analysis-workflow-using-pygwalker-and-feature-engineered-data

How to Build an Advanced, Interactive Exploratory Data Analysis Workflow Using PyGWalker and Feature-Engineered Data

2+ day, 4+ hour ago  (225+ words) We set up a clean and reproducible Colab environment by installing all required dependencies for interactive EDA. We load the Titanic dataset and perform an initial sanity check to understand its raw structure and scale. It establishes a stable foundation…...