#AI News

1841 articles

Victoria Bazir Victoria Bazir AI models
8 months ago

ChatGPT Sparks Healthcare Debate After Viral MRI Review Claim

A viral post claiming ChatGPT helped review an MRI scan has reignited debate about AI's role in healthcare, highlighting growing public reliance on AI tools for medical guidance despite lacking regulatory approval.

Saad Ullah Saad Ullah Industry
8 months ago

Manus Rolls Out Design View Powered by Google's Nano Banana Pro

Manus just dropped Design View, a visual design extension that lets you generate and tweak designs in one smooth workflow. It's live for everyone now and runs on Google's Nano Banana Pro tech.

Eseandre Mordi Eseandre Mordi Benchmarks
8 months ago

MiniMax M2.1 vs GLM-4.7: 74.0 vs 73.8 on SWE-bench

First benchmark results put MiniMax M2.1 and GLM-4.7 neck-and-neck on SWE-bench Verified, though MiniMax pulls ahead on Terminal Bench 2.0. Keep in mind these are vendor-reported numbers, not independent tests.

Usman Salis Usman Salis AI models
8 months ago

GLM 4.7 Achieves 73.8% on SWE-bench, Leading Open-Source AI in 2025

GLM 4.7 closes out 2025 as the strongest open-source AI model, delivering benchmark scores that put it within striking distance of premium closed-source systems while maintaining cost efficiency and accessibility for developers.

Saad Ullah Saad Ullah AI models
8 months ago

Humanoid Robot Completes 6 Household Tasks as Pi 0.6 Tackles Moravec Paradox

A new demonstration shows a humanoid robot performing everyday household chores that humans typically do automatically. The results highlight progress in addressing the Moravec paradox in robotics.

Eseandre Mordi Eseandre Mordi AI models
8 months ago

xAI Launches Grok Collections API With $2.50 Per 1,000 Searches Pricing

xAI introduced the Grok Collections API, a built-in retrieval system for handling large document datasets. Early benchmarks indicate it outperforms both Gemini 3 Pro and GPT-5.1 in retrieval accuracy.

Peter Smith Peter Smith AI models
8 months ago

GLM-4.7 Delivers 128K Context Performance Across 8 Major Benchmarks

Z.ai's latest open-source model shows measurable gains in coding accuracy and reasoning tasks

Usman Salis Usman Salis Markets
8 months ago

AAOI Stock Jumps 13% as Applied Optoelectronics Lands First 800G Hyperscaler Deal

Applied Optoelectronics shares surged over 13% after securing its first major order for 800G data-center transceivers, signaling growing demand for optical networking in AI infrastructure.

Saad Ullah Saad Ullah AI models
8 months ago

Google's Titans and MIRAS Could Let AI Models Update Memory While Running

Google Research unveiled a new approach combining Titans architecture with MIRAS that lets AI models update their long-term memory during execution, potentially moving beyond traditional context window limitations.

Eseandre Mordi Eseandre Mordi Industry
8 months ago

Gemini 3 Flash Creates 3D Racing Game in Just 1 Minute

Early users are impressed by Gemini 3 Flash's ability to quickly generate interactive 3D applications from simple text prompts. The model reportedly works faster and feels more accessible than previous Gemini versions.

Sergey Diakov Sergey Diakov Trends
8 months ago

AI Could Match Human Reasoning in 5 Years as Performance Jumps Exponentially

New analysis of AI development trends reveals exponential performance gains across key benchmarks, with planning and reasoning capabilities potentially reaching human-level within five years if current growth continues.

Alex Dudov Alex Dudov Industry
8 months ago

Adobe Researchers Discover How Spatial Vision Cuts AI Training Time by 40%

Adobe-affiliated researchers reveal that spatial structure, not teacher model accuracy, drives faster generative AI training. A minimal update called iREPA delivers quicker convergence and stronger results across models.

Usman Salis Usman Salis AI models
8 months ago

ChatGPT Holds 74% Search Share as Gemini Climbs to 15% in 2025

Similarweb data shows shifts in search traffic among leading generative AI websites in 2025, with Gemini gaining share while ChatGPT remains the dominant destination.

Peter Smith Peter Smith AI models
8 months ago

Humanoid Robot Dance Shows Hit 100+ Performances as Motion Programming Goes Mainstream

Programmed choreography has transformed once-groundbreaking robot performances into routine demonstrations, with hundreds of synchronized shows now executed with near-perfect consistency.

Saad Ullah Saad Ullah AI models
8 months ago

AgiBOT Unveils Genie2: New Wheeled Humanoid After Hitting 5,000-Unit Production Milestone

AgiBOT has introduced Genie2, a wheeled humanoid robot designed to automate picking, navigation, and shelf management in retail fulfillment environments.

Artem Voloskovets Artem Voloskovets Industry
8 months ago

Ant Group Scales Diffusion Language Models to 100B Parameters with LLaDA2.0

Ant Group released LLaDA2.0, a framework for scaling diffusion language models up to 100 billion parameters. The method converts existing auto-regressive models while boosting efficiency and performance.

Peter Smith Peter Smith Benchmarks
8 months ago

DeepWriter Tops HLE Rankings with 50.91 Score Among Agentic AI Systems

Benchmark results show DeepWriter hitting the highest HLE score among leading agentic AI tools. The data reveals significant differences in long-form reasoning and autonomous performance capabilities.

Saad Ullah Saad Ullah Industry
8 months ago

Nvidia's Nemotron-Cascade Pushes AI Reasoning to 75% Accuracy

Nvidia launched Nemotron-Cascade, a reinforcement learning framework designed specifically for reasoning-focused AI models. Testing shows consistent accuracy improvements through staged training phases.

Marina Lyubimova Marina Lyubimova AI models
8 months ago

Claude Opus 4.5 Reaches 5-Hour Success Threshold in New AI Benchmark

Fresh benchmark data reveals Claude Opus 4.5 leading the pack in long-duration AI task performance, sustaining successful execution for nearly five hours at 50% success probability.

Saad Ullah Saad Ullah Industry
8 months ago

OpenAI's Compute Margins Jump to Nearly 70% as Efficiency Gains Take Hold

OpenAI's profit margins on compute costs for paying customers have surged to nearly 70% by October, driven by cheaper infrastructure and smarter inference technology.

Posts feed
Anthropic Opens MHS Preview to Connect AI Agents With Physical Hardware
22 hours ago
OpenAI's $122 Billion Superapp Strategy Signals a Major AI Platform Shift
4 months ago
BitVLA Shrinks Robot AI by 11x While Boosting Speed 4.4x
4 months ago
LFM2.5-350M Debuts With Strong Benchmark Results Across 6 Evaluation Suites
4 months ago
Grok 4.20 Multi-Agent Beta Ranks #7 in Search Arena Leaderboard
4 months ago
Claude Code Leak Reveals 4 Key Features Powering Autonomous AI Systems
4 months ago
RISE Robot Framework Boosts Task Success Rates by 35% Using AI-Driven Imagination
4 months ago
Multi-Agent AI Systems Face 15 Emerging Social Risks, New Study Finds
4 months ago
Mistral AI Raises $830M to Build Paris AI Data Center
4 months ago
OpenAI Codex Adds Claude Code CLI Integration in Cross-Platform Move
4 months ago
Follow us on Twitter

Aigazine is an independent publication covering artificial intelligence — new models, research, the companies building them, and the industries they reshape. Published from Tbilisi, Georgia (0179, Tbilisi City, Vake District, 49 Besarion Zhghenti Street, VAT 305786600).

© 2026 Aigazine

We use cookies to improve your experience on our site and to show you relevant advertising. To find our more, read our privacy policy and cookie policy