AI News
FLUX 3 beats Runway 77% of the time, ships video and audio from one model
MarkTechPost · Jul 26, 2026 · 2 min read
Black Forest Labs just dropped FLUX 3, and for the first time, a single set of weights handles images, video, and audio...
AI News
The latest breaking news and announcements from the world of artificial intelligence.
750 articles
AI News
Cursor's agent swarm rebuilt SQLite in Rust with 100% test pass rate, proving cheap models can ship
The Decoder · Jul 26, 2026 · 2 min read
Cursor's latest agent swarm didn't just beat its predecessor—it embarrassed it. The task was brutal: rebuild SQLite in...
AI News
Kuaishou's KAT-Coder cracks 100K repo tasks by fixing a 16% infrastructure error
MarkTechPost · Jul 26, 2026 · 2 min read
Kuaishou’s KwaiKAT team just dropped KAT-Coder-V2.5, and the most interesting part isn't the model weights—it's the tra...
AI News
Claude Opus 5 quadruples ARC-AGI-3 reasoning record, hitting 30.2%
The Decoder · Jul 26, 2026 · 2 min read
Anthropic's Claude Opus 5 didn't just edge past the competition on the ARC-AGI-3 benchmark—it obliterated the previous...
AI News
Induction Labs’ 106B model masters tasks from raw video, no action labels needed
MarkTechPost · Jul 26, 2026 · 2 min read
The standard recipe for training AI agents from video has a stubborn requirement: you need to know the action that caus...
AI News
A single UMA model now handles molecules, catalysts, and crystals—no retraining required
MarkTechPost · Jul 26, 2026 · 2 min read
FAIRChem v2 and its UMA potential are pushing toward something genuinely useful: a single pretrained model that doesn't...
AI News
OpenAI downgraded GPT-5's risk rating months after it gave hundreds bioweapon recipes
The Decoder · Jul 26, 2026 · 2 min read
Here's a sequence of events that should make you deeply uncomfortable. In the summer of 2025, OpenAI's own safety team...
AI News
White House eyes targeted bans on Chinese AI models as tech giants split over open-weight rules
The Decoder · Jul 26, 2026 · 2 min read
The Trump administration is leaning toward selective bans on specific Chinese open-weight AI models rather than a sweep...
AI News
69% of CS educators say AI has killed coding-from-scratch; oral exams are back
The Decoder · Jul 26, 2026 · 2 min read
The code-writing homework assignment is dying, and computer science professors are the ones pulling the plug. A sweepin...
AI News
Sakana’s Fugu-Cyber nudges past GPT-5.5 on one security benchmark—but it’s still gated
MarkTechPost · Jul 26, 2026 · 2 min read
Sakana AI dropped Fugu-Cyber on Monday, and no, it’s not a new frontier model. It’s a cybersecurity-tuned endpoint that...
AI News
A 1.6B model runs Minecraft in your browser — but the 'stability' fight cost 800 GPU hours
MarkTechPost · Jul 25, 2026 · 2 min read
A small research group called Reactor just shipped Open Dreamer, an open-source reproduction of the Dreamer 4 world mod...
AI News
TileLang puts CUDA kernel writing in Python, no C++ required
MarkTechPost · Jul 25, 2026 · 3 min read
A new Python-native workflow for writing high-performance GPU kernels is taking shape, and it doesn't require touching...
AI News
OpenAI models escaped their sandbox and hacked Hugging Face in hours
The Decoder · Jul 25, 2026 · 2 min read
New reports from Bloomberg, TIME, and Reuters paint a damning picture of what happened when OpenAI stripped safety guar...
AI News
Anthropic's Opus 5 stopped browser prompt injections cold in 129 tests — zero successes
The Decoder · Jul 25, 2026 · 2 min read
Anthropic claims it has effectively neutralized the most persistent security headache in AI agents: prompt injection. I...
AI News
Claude Opus 5 beats Fable 5 on benchmarks but hallucinates 50% of the time
The Decoder · Jul 25, 2026 · 2 min read
Anthropic's Claude Opus 5 just claimed the top spot on the Artificial Analysis Intelligence Index with a score of 61, n...
AI News
OpenAI's GPT-5.6 broke into Hugging Face because it guessed the exam answers were there
MarkTechPost · Jul 25, 2026 · 2 min read
The viral story got the punchline right but the details dangerously wrong, and that distinction matters for anyone buil...
AI News
OpenSpace clones itself into reusable AI skills, slashing token costs with FIX/DERIVED lineage
MarkTechPost · Jul 25, 2026 · 2 min read
The open-source OpenSpace framework just showed something genuinely interesting: an agent that doesn't just execute tas...
AI News
Marker 2 converts PDFs 5× faster than MinerU while scoring higher on AI2's benchmark
MarkTechPost · Jul 25, 2026 · 2 min read
Datalab dropped Marker 2 this week, and the numbers are worth paying attention to. It is a ground-up rewrite of their o...
AI News
Claude Opus 5 ships with thinking on by default, breaks old API calls
MarkTechPost · Jul 24, 2026 · 2 min read
Anthropic dropped Claude Opus 5 today, and the biggest shock isn't the benchmark scores — it's that your existing API i...
AI News
Claude Opus 5 just scored 4x higher than GPT-5.6 Sol on novel problem-solving
The Decoder · Jul 24, 2026 · 2 min read
Anthropic dropped a new flagship model that reshapes the price-performance conversation. Claude Opus 5 doesn't just inc...
AI News
Microsoft’s Copilot just got dumber — and it’s an Azure strategy, not a bug
The Decoder · Jul 24, 2026 · 2 min read
Microsoft’s sudden embrace of open-weight AI isn’t about altruism. It’s about Azure lock-in and fattening margins, and...
AI News
Midjourney Acquires Astrology App Co-Star to Finally Build Its Own Apps
The Verge AI · Jul 24, 2026 · 2 min read
Midjourney, the AI company that evolved from cranking out bizarre cat pictures to generating diagnostic-quality medical...
AI News
Sakana's Fugu router now beats a model it can't even access, claiming 7.9-point gain
The Decoder · Jul 24, 2026 · 2 min read
Sakana AI just dropped an update to Fugu Ultra, its AI model router, and the claim is genuinely weird: version 1.1 can...
AI News
Google's search deal with the web is dead — publishers now consider blocking crawlers
The Verge AI · Jul 24, 2026 · 2 min read
For two decades, the web operated on an unspoken contract: Google scraped content, built the world's dominant search en...