Skip to content
Tech News
clear
Topics: Today This Week This Month This Year

Google launches Gemini Omni 1.1 Flash with extended video generation via Gemini API

Google has released Gemini Omni 1.1 Flash, adding creative controls and generative video features aimed at developers building media and editing tools through the Gemini API in Google AI Studio. The update introduces scene extension, letting users continue existing videos using up to 10 seconds of prior context instead of just the final second, and extend clips in 10-second increments up to 40 seconds total.

YMTC targets NAND market leadership by 2027 as it files for $4.9B Shanghai IPO

Chinese chipmaker YMTC has told investors it intends to become the world's biggest NAND flash producer by the end of 2027, surpassing Samsung and SK hynix. The Wuhan-based firm recently filed to raise 33 billion yuan ($4.9 billion) on Shanghai's STAR Market, with most funds going toward upgrading production lines and expanding research and development.

Z.ai unmasks mystery model Ox Alpha as open-weight GLM-5.3-Flash

An unidentified free model called Ox Alpha appeared on OpenRouter and quickly drew massive usage from developers who couldn't determine its origin, sparking days of speculation about which lab built it. Z.ai revealed the model was actually its own GLM-5.3-Flash, running entirely on Chinese chips, and priced at 15 cents per million input tokens and 50 cents per million output tokens, with a launch discount cutting that in half through September 9.

LM Studio brings Z.ai's GLM-5.3-Flash to Bionic with vision and 1M-token context

LM Studio has integrated Z.ai's newly released GLM-5.3-Flash model into its Bionic agent platform, adding multimodal image input and a 1 million-token context window. The company says the model outperforms its predecessor, GLM-5.2, while costing roughly 9-10 times less to run, and it's served from US-based servers with zero-data-retention enabled by default.

Glacial collapse, not earthquake, triggered fatal Nepal-Tibet flash flood

A massive ice and rock collapse in the mountains along the Nepal-Tibet border unleashed a torrent of mud, rock and water down the Bhotekoshi River on Wednesday, killing around 350 people with more than 1,000 still missing. The US Geological Survey confirmed the landslide itself generated a 5.2-magnitude seismic signal, ruling out earlier theories that an earthquake or a glacial lake burst had caused the disaster.

Zhipu's GLM-5.3-Flash scores 57 on Intelligence Index at low token pricing

GLM-5.3-Flash is a text-only model with a 400k token context window that scored 57 on the Artificial Analysis Intelligence Index, far above the comparable median of 18. It is priced at $0.15 per 1M input tokens and $0.50 per 1M output tokens, both below the median rates for similar models, with the full benchmark run costing $138.02.

Nokia 300 Charge turns a feature phone into a 10W powerbank

Nokia's new 300 Charge dumb phone packs a 3,700mAh battery capable of 37 hours of talk time, but its standout feature is a reverse-charging 'powerbank mode' that outputs 10W via a bundled 4-in-1 cable with Micro USB, USB-C and USB-A adapters. The phone also includes a bright LED flashlight, a 3.5mm headphone jack, and basic specs like a Unisoc SC6531E chipset, 4GB RAM/storage, a QVGA camera and a 2.4-inch display.

Pixel 11 Pro's LED flash measures weaker than Pixel 10 Pro in early tests

Reddit users and Android Authority's own lux measurements indicate the Pixel 11 Pro's camera flash and flashlight are noticeably dimmer than the Pixel 10 Pro's, with one test showing 89 lux versus 123 lux. The likely culprit is a physical diffuser added around the camera visor to accommodate Google's new HiLight ambient LED array, which appears to soften the main flash's beam.

Hot Chips 2026 tutorial details how High Bandwidth Flash could reshape ML memory

At Hot Chips 2026, researchers Anurag Agarwal and Radhakrishna Giduthuri presented a tutorial on High Bandwidth Flash (HBF), a package-integrated flash memory technology built like HBM but functioning more like an SSD. Since no HBF hardware exists yet, the talk relied on simulations and projections, examining how software—especially inference frameworks like vLLM—would need to be redesigned to exploit HBF's large capacity despite its coarse-grained, storage-like access pattern.

Today's top topics: donald trump artificial intelligence united nations
View all today's topics →