AI Research & Science
Peer-reviewed breakthroughs, university studies, and lab discoveries — explained in plain English. AI Maestro tracks the frontiers of machine learning research, neuroscience meets AI, and the science driving the next wave of intelligent systems.
Claude Haiku 5.5 arrives with massive price cuts proving the AI pricing arms race is far from over
Claude Haiku 5.5 arrives with massive price cuts proving the AI pricing arms race is far from over Anthropic has released Claude…
Top stories
Mistral Large 4
yesterday
Researchers stretch LeCun’s JEPA AI into a universal world model that works from physics to biology
yesterdayReflection AI Introduces Beam: A 501B Open-Weight MoE Model With 23B Active Parameters for Coding and Agentic Workloads
2d ago
Reflection debuts Beam, an open-weight AI model to rival Chinese models at lower compute cost
2d agoMore ai research & science
Meet NeuroVFM: A New Neuroimaging Foundation Model Trained With Vol-JEPA on Uncurated Clinical MRI and CT Volumes
Frontier models usually train on public internet data, but clinical neuroimaging is rarely available…
13 Jul 2026
Scientists’ Side Hustle? Using AI and Quantum Computing to Generate New Peptides
Researchers at the Technical University of Denmark have demonstrated that a printer-sized quantum computer…
12 Jul 2026
Scientists Solve Mystery of Bizarre ‘Alien Megastructure’ Star
Researchers have identified a giant planet orbiting Tabby’s Star, offering a natural explanation for…
11 Jul 2026
Ant Group’s Robbyant Unveils LingBot-VA 2.0: A Causal Video-Action Model Built Natively for Physical AI
Robbyant, the embodied AI unit within Ant Group, has released LingBot-VA 2.0. This is…
11 Jul 2026
OpenAI’s GPT-5.6 Sol autonomously post-trained the smaller Luna model with a “fairly underspecified prompt”
OpenAI’s new GPT-5.6 Sol model independently post-trained the smaller Luna model using a prompt…
10 Jul 2026
The new GPT-5.6 family: Luna, Terra, Sol
OpenAI launched GPT-5.6 this morning in three configurations: Luna, Terra, and Sol. Pricing and…
9 Jul 2026
GPT-5.6 Sol nearly matches Fable 5 on aggregated benchmarks at one-third the cost
OpenAI has launched GPT-5.6 Sol, a flagship model that scores 59 points on the…
9 Jul 2026
OpenAI finds roughly 30 percent of popular AI coding test is broken
OpenAI has stopped recommending SWE-Bench Pro as a standard for testing AI coding skills…
9 Jul 2026
Grok 4.5 is so cheap compared to Fable 5 and GPT 5.5 that benchmark gaps may not matter much
Grok 4.5 is so cheap compared to Fable 5 and GPT 5.5 that benchmark…
8 Jul 2026