Love seeing what happens when open weights meet great engineering!
In a new benchmark, @AirbnbEng took Gemma 4 E2B for policy classification and served it on a single @NVIDIA RTX Pro 6000 GPU:
🎯 Quality: Jumped from 0.563 to 0.902 max F1, matching top frontier APIs
⚡ Speed: 35ms median latency (85ms p99) vs. ~900ms for frontier models
💸 Cost: ~1% of frontier API spend at high volume
Results like these come down to @AirbnbEng's thoughtful evals, domain data, and knowing exactly when a specialized small model beats a massive prompt. Grateful Gemma could be part of the stack!
Por qué todas las miradas están puestas en la salida a bolsa de Anthropic
La empresa de inteligencia artificial creadora del asistente Claude debutó en Wall Street con un precio de 111,50 dólares por acción, lo que supone una subida del 34 % respecto al precio de salida de 83 dólares por título. Aun así, algunos analistas advierten de que persiste cierto escepticismo en los mercados sobre el futuro del sector de la inteligencia artificial, debido al elevado coste de las inversiones necesarias, los riesgos asociados a esta tecnología y la creciente competencia entre las empresas del sector.
Anthropic’s compute spending is increasing by 4x per year, outpacing OpenAI’s 3x.
Per its S-1, Anthropic spent $10 billion on compute and infrastructure in Q3 2026 alone, or $40 billion annualized, compared to $433 million in all of 2023.
Google continues to accelerate its release cadence with Gemini 4 Neon, delivering a small Intelligence gain over Gemini 4 Carbon at a ~17% lower Cost per Task
@GoogleDeepMind has released Gemini 4 Neon, its fourth model in just seven weeks. Neon (xhigh) scores 61 on the Artificial Analysis Intelligence Index, 1 point ahead of Gemini 4 Carbon (high) and 3rd overall, behind GPT-6.2 Astra (max) and Claude Fable 5.2 (max). At high effort, Neon matches Carbon (high) at less than half the Cost per Task using Carbon’s standard (non-discounted) pricing.
Key takeaways:
➤ Slight improvement in Intelligence for lower Cost and Time per Task: Neon (xhigh) scores 61 on the Intelligence Index, 1 point ahead of Carbon (high), with small improvements in agentic knowledge work, visual reasoning and coding. It costs $6.44 per task, ~17% less than Carbon ($7.76 at non-discounted pricing), and takes ~21 minutes per task, ~40% less than Carbon (~36 minutes). Neon uses ~15% more output tokens per task, so the lower cost comes purely from its 25% lower token prices ($3/$15 vs $4/$20 per 1M input/output tokens), and the lower time from its higher output speed (~110 vs ~57 output tokens per second)
➤ Competitive with Claude Opus 5.5 and Claude Fable 5.2 on coding agents: With Antigravity CLI, Neon (xhigh) scores 59 on the Artificial Analysis Coding Agent Index, 3 points ahead of Carbon (high, 55). It scores in line with Claude Code with Claude Fable 5.2 (xhigh, 58) and Claude Opus 5.5 (max, 58) at a much lower Cost per Task. Gemini 4 Neon in Antigravity has an average Cost per Task of $6.86, compared to $12.61 for Claude Opus 5.5 (max) and $15.47 for Claude Fable 5.2 (xhigh). Neon is available in Antigravity CLI for Pro and Ultra subscribers
➤ Significant improvements in visual reasoning and output presentation quality: In GDP.pdf, Gemini 4 Neon (high) scores 33.4%, 5 points above Carbon (high, 28.4%) and in line with the highest scores measured, while Neon (xhigh) scores 28.0%. On AA-Briefcase, Neon (xhigh) scores 1700 Elo overall and leads on Presentation Quality with 1771 Elo, 45 points ahead of the previous leader Claude Sonnet 5.5 (max, 1726). On Google’s Multimodal Knowledge Agent evaluation (MKA), Neon reaches a Climb WSR of 56%, up 2 points from Carbon and on par with the leader, Claude Fable 5.2 (max, 57%)
Key model details:
➤ Context Window: 1M tokens
➤ Multimodality: Text, image, video, and speech input, with text output
➤ Availability: Gemini 4 Neon is publicly available at launch through Google’s API. Pro and Ultra subscribers can also access it in Antigravity and the Gemini app
➤ Effort levels: Gemini 4 Neon offers four effort levels (low, medium, high, and xhigh) - xhigh was introduced to Gemini models in the recent Hydrogen release
➤ Pricing: $3/$15 per 1M input/output tokens, 50% above Gemini 4 Carbon’s currently discounted price of $2/$10 (which ends on November 30), and 25% below its standard $4/$20. Cached input tokens receive the same 95% discount ($0.15)
🔴 ¡GOOGLE ANUNCIA GEMINI 4 NEON!
Y con este ya son 4 los elementos químicos presentados en las últimas semanas (Argon, Carbon, Helio y Neon)
Este modelo se presenta como una gama media-alta, o lo que es lo mismo el equivalente a los modelos Opus y Sol.
Over the past week, I’ve been overwhelmed by the love and support of friends, industry colleagues, and people I've never met. Thank you to everyone who has shown kindness to my family and me. It means the world to me.
I came to the United States 53 years ago, and becoming an American citizen is one of the greatest gifts and proudest moments of my life. NVIDIA has been my life’s work. Building it into the company it is today alongside my colleagues has been a privilege and an honor. Nowhere but in America could a company like NVIDIA be possible. I love our country, and I love NVIDIA.
I did not live up to the standards expected of everyone at NVIDIA, and I have made serious mistakes in my personal life. I created problems for the Board that I deeply regret. I own my mistakes completely, and I am committed to cooperating with the Board’s independent investigation.
Since my departure, I’ve seen an outpouring of calls from NVIDIA’s employees, investors, business partners, and customers for me to return as CEO. Many continue to reach out to me directly as the company navigates complex technology, market, competitive, and geopolitical dynamics. NVIDIA is a once-in-a-generation company. If the Board of Directors decides that my return is in the best interest of our shareholders, our employees, the company, and America, I would be honored to answer these calls and return as CEO.
Thank you again to everyone who has shared their good wishes. I am profoundly humbled by your support. To my family: I love you, and I will spend the rest of my life being worthy of yours.
Jensen
Perception generally available! It orchestrates blue, red and green team agents through playbooks, working together in a closed loop to secure your systems. Perception helps security teams move at machine speed, driving outcomes not tasks and automating while staying in control.
New AI models announced at Microsoft Ignite! 🚀
Meet Microsoft Decision-1.1, MAI-Code-2, and Cyber-2-Flash, expanding what's possible with AI for decision-making, coding, and security. #MSIgnite
Want to learn more?
Explore our new learning playlist and discover how to build with Microsoft models: https://t.co/KMn4d6RJNc
La situación hoy pinta tal que así: caidas y ciberataques en múltiples servicios.
Y ante todos estos titulares, junto a otras caídas de servicios claves de días anteriores, es normal pensar si la IA tiene algo que ver. Lo que se sabe es que algunos de estos casos (NSA, iCloud, etc), con los indicios que tienen, no parecería tratarse de ciberataques como tal descartándose por tanto la implicación de la IA.
Aún así, es una fotografía de una realidad hacia la que nos movemos a ritmo rápido.
Must-read papers of the week
▪️ World In-Context Learning for Agentic Robot Manipulation
▪️ Don’t Just Trust Your Memory
▪️ Is Your Robot’s Memory Just a Clock? Temporal Shortcuts in Memory-Augmented VLA Policies
▪️ Unbundling Multimodal Vision Encoding
▪️ V-JEPA-LM
▪️ Learning Heterogeneous Tactile Representation for General Dexterous Manipulation
▪️ Training Reasoning Models in Simulation to Natively Use Real Robots
▪️ Embodied First Exam
▪️ Continual-ARC: Measuring Skill Acquisition over a Working Lifetime
▪️ Reframe: Observation-Aligned Multi-Teacher On-Policy Distillation for Agentic Visual Reasoning
▪️ MemInStream: Memorizing or Merely Recognizing? Rethinking Streaming Video Memory Evaluation
▪️ Infrathon: Benchmarking Language Agents for Real-World, Long-Horizon AI Infra Optimization
▪️ Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents
▪️ GTAGym: AAA Game Tasks Environment for Evaluation and Reinforcement Learning
Explore these to keep up with main AI trends. Here’s also the full list of outstanding papers + links, and our weekly AI news digest: https://t.co/XJLza8hzI8
Just announced at #MSIgnite: the GitHub Copilot app now has Azure DevOps integration ☁️
Clone an Azure DevOps repo, and see the pull request through to completion right in the app:
• track its status in the sidebar
• click Fix on a review comment to have an agent address it
• or turn on agent merge to let an agent fix CI failures and complete the pull request once your Azure DevOps checks and policies pass.
Now in public preview ⬇️
https://t.co/sOwrCNPSYc
Nemotron 3.5 Transcribe takes #1 on @ArtificialAnlys’ new Diarization Index! 💚
Thanks to the team for testing how well models identify who said what in conversations with 2–8 speakers.
Dario Amodei on open source models:
“If the path continues, I think we could get to a very dangerous place.”
“When you control a model and you're deploying it, you have the ability to monitor its usage.”
“You can revoke a user's access. You can change what the model is willing to do.”
“When a model is released in an uncontrolled manner, there's no ability to do that. It's entirely out of your hands.”
“We should confront the advocates of this with that problem and challenge them to solve it.”
He clearly doesn't believe Intelligence should be in the hand of the masses and Sam Altman also saying "RAM" shouldn't be accessible to everyone indefinitely agreed with him.
GPT-6.2 Astra is the new leader in AA-Robotics v0.1, scoring 79, up 10 points on GPT-6 Astra and ahead of GPT-6.2 Sol (78), Claude Fable 5.2 (75) and Gemini 4 Hydrogen (58)
Controlling a real robot arm through Codex, it passed six of our eight physical tasks, partly completed the pouring liquid, and failed only to measure the box.
New models built for coding and cybersecurity, with cost-performance in focus. Meet Code 2 for repo-wide reasoning, Code 2 Flash for fast, high-volume coding, and Cyber 2 Flash, our most capable security model yet. #MSIgnite
🔴 ¡OPENAI ANUNCIA GPT 6.2 ASTRA!
Se ha hecho de rogar, pero por fin tenemos aquí la nueva generación GPT 6.2 de modelos tanto para Astra como para Sol.
Os amplío la noticia 🧵👇
Today we’re announcing the general availability of Microsoft-Decision-1.1, with greater intelligence, better calibration and more than twice the speed of our first model, at a low cost.
https://t.co/V0VX0EqgAF
Looped Transformers are low-key fascinating. First of all, you need to understand what travels around the loop and the 3 core things you set for the architecture.
A Looped Transformer runs the same Transformer block (or stack) for multiple passes, feeding each output back in as the next input.
▪️ What travels around this loop is the hidden state:
- Token embeddings form the initial hidden representation.
- The shared stack updates that representation via attention and feed-forward operations.
- Its output goes back to the stack for another pass, using an updated hidden state. Some designs also reuse the input, add time-step embeddings, or run multiple residual streams.
- After the chosen number of passes, the output head converts the selected representation into token scores.
That's all to win additional computation without an additional set of block weights for every pass.
Although the weights stay the same during inference, changing inputs let the same block produce different updates. During decoding it attends via the KV cache, during prefill it updates masked tokens in parallel.
A model can loop its entire stack or repeat only a middle stack between ordinary input and output blocks.
Note that repeating the stack doesn't require generating a new text token between passes.
▪️ So there are 3 basic things you need to identify for the specific Looped Transformer architecture:
1. The repeated blocks
2. The state they receive
3. The rule that controls their execution - a fixed number of passes, a user-selected count, or an adaptive rule
If you want to nerd out on the rest, like adaptive depth, what happens to the KV-cache and hardware, 2026 breakthroughs, and when another pass is worth the cost - we put together a full guide over here: https://t.co/dFzHjmH5RO