Skip to main content

Author: Felipe Santos

"Artificial intelligence can process the world in milliseconds, but only the human heart can give meaning to every second lived" - Mr. Santos
Developers Expand Apps Across iPhone Duo's Large Inner Display

Developers Expand Apps Across iPhone Duo’s Large Inner Display

Apple is showing how apps from streaming services, productivity tools, games, and local businesses are being adapted for iPhone Duo, a folding device whose inner display gives apps a larger canvas. The apps adjust their layouts to each position the device can take: fully open, partially folded, propped up on a table, or closed. Entertainment and Sports Netflix lets members scroll through Clips on the outer display, then open iPhone Duo to keep watching on the large nano-texture inner display. Clips switch seamlessly between vertical and horizontal views. Prime Video and Paramount+ are...

Continue reading

Google Cloud adds three OpenAI GPT-6 models

Google Cloud adds three OpenAI GPT-6 models

OpenAI's GPT-6 Astra, GPT-6.1 Sol, and GPT-6 Luna are joining Google Cloud's model garden, making Google Cloud the only hyperscale cloud provider with access to all three leading closed-source LLMs. The models sit alongside more than 200 first-party, third-party, and open models on a unified enterprise AI platform. Google Cloud frames the move as an extension of its long-standing emphasis on choice, enterprise-grade governance, and price-performance, and as a way for customers to run agentic workloads across the models and infrastructure that best balance cost, capability, and...

Continue reading

Eight Spark nodes pool 1 TB of memory

Eight Spark nodes pool 1 TB of memory

NVIDIA announced official support for eight-node DGX Spark clusters at GTC Berlin, scaling unified memory to 1 TB. Each DGX Spark provides 128 GB of unified memory, so four nodes reach 512 GB and eight nodes reach 1 TB. That capacity opens the door to models that previously wouldn't start on a four-node system, including the full GLM 5.3 (753B) model from Z.ai. The benchmarks show that eight nodes deliver the memory headroom and near-linear training scaling, while the best network topology depends on the workload. Support for eight-node clusters in NVIDIA Sync is coming by the end of the...

Continue reading

CORSAIR unveils two Cyberpunk 2077 gaming peripheral collections

CORSAIR unveils two Cyberpunk 2077 gaming peripheral collections

CORSAIR has announced two officially licensed gaming peripheral collections made with CD PROJEKT RED, with pre-orders starting October 22, 2026. The first, the CORSAIR x Cyberpunk 2077: Arasaka White Collection, brings back last year's Arasaka Edition lineup in the megacorp's white and red look. The second, the all-new CORSAIR x Cyberpunk: Edgerunners 2 Collection, ties into the new anime series from Studio TRIGGER and CD PROJEKT RED, which is streaming on Netflix. Last December, CORSAIR and CD PROJEKT RED marked the fifth anniversary of Cyberpunk 2077 with the black Arasaka Edition...

Continue reading

Samsung Boosts PRO Plus Cards to 180 MB/s

Samsung Boosts PRO Plus Cards to 180 MB/s

Samsung Electronics America has upgraded its PRO Plus memory cards, which are aimed at professional and enthusiast photographers, videographers, and content creators. The new Samsung PRO Plus microSD and full-size SD cards reach read speeds of up to 180 MB/s and write speeds of up to 130 MB/s. Those figures are increases of 12 percent (read) and 8 percent (write) over their predecessors. Samsung says the cards are built for professional use, letting users capture 4K ultra-high definition (UHD) and Full HD (FHD) video and photos. They also carry a V30 video speed class rating, which Samsung...

Continue reading

Samsung packs 200 million pixels into 1/1.3-inch sensor

Samsung packs 200 million pixels into 1/1.3-inch sensor

On January 18, 2023, Samsung Electronics introduced the ISOCELL HP2, a 200-megapixel (MP) image sensor with improved pixel technology and full-well capacity. Samsung aims it at “tomorrow's premium smartphones,” where it is meant to deliver “stunning mobile images.” The sensor's main technical point is its packing density within a familiar size. The ISOCELL HP2 uses 200 million 0.6-micrometer (μm) pixels in a 1/1.3-inch optical format. Samsung notes that this sensor size is already widely used in 108MP main smartphone cameras, so manufacturers can move to much higher...

Continue reading

NVIDIA Cosmos delivers up to 2x faster inference

NVIDIA Cosmos delivers up to 2x faster inference

NVIDIA has expanded its Cosmos platform with new training tools, simulation checkpoints and Cosmos Data Factory blueprints, aimed at developers building world models, which help physical AI systems understand and simulate the real world. NVIDIA says the platform ranks in the top five on six different open model leaderboards. It pairs the frontier open models with data, training and optimization tools and recipes so builders can scale training data, adapt models, test behavior and optimize performance before deployment. .abn-galeria .uk-slidenav, .abn-galeria...

Continue reading

Razer's Kraken Kitty V3 lights up 16.8 million colors

Razer’s Kraken Kitty V3 lights up 16.8 million colors

Razer has announced the Kraken Kitty V3, a wired RGB headset for streamers who want their gear to show their style and engage their community. It builds on the Kraken Kitty line, and its stream-reactive kitty ears and earcups, powered by Razer Chroma RGB, are the main design feature. The headset also includes the Razer HyperClear Cardioid Mic and immersive audio. Razer says it is designed to move from streaming setup to cosplay accessory. The headset uses plug-and-play USB Type-C connectivity for compatible PC and mobile devices, so creators can get connected and go live in seconds. Hybrid...

Continue reading

Agents recall past sessions 8x cheaper than handoffs

Agents recall past sessions 8x cheaper than handoffs

Coding agents forget everything when a session ends, so each new agent on each new machine starts from zero. Hugging Face's funes is an open-source memory layer that indexes the sessions already on your machine and lets Claude Code, Codex, pi, and Hermes recall past decisions, rationale, and dead ends. In the team's benchmark, recall was 8x cheaper than a written handoff on one task and 4x cheaper on the other. The article builds on an earlier argument that coding agents already produce the record we keep losing: as they search, try approaches, hit errors, and change direction, they leave a...

Continue reading