Omdia expects wider-screen foldable phones to drive a rebound in AMOLED panel shipments, with volumes rising 44% year over year to 29 million panels in 2026. Wide-screen designs are forecast to account for 44% of foldable panel shipments this year and more than 60% in 2027, as Chinese phone makers adopt the format. The firm also expects Samsung Display’s market share to reach 66% in 2026, citing its exclusive supply of panels for Samsung’s Galaxy Fold8 and Apple’s iPhone Duo. Panel and phone shipments are tracked separately and can fall in different quarters.

Omdia: Memory > DRAM industry revenue is forecast to reach $372 billion in 2026, representing a staggering 147% year over year growth, while NAND revenue is projected to hit a record $300 billion, up 311% year over year. > Memory accounts for approximately 10-15% of total hyperscaler capex, but when factoring in higher DDR pricing and HMB4 driven density expansion, memory spending alone could explain roughly 15-25%. > Average DRAM content per server box is projected to grow from 1,032GB in 2025 to 1,432GB in 2026, a 39% year over year increase driven by AI inference workloads that demand exponentially more memory. > The DRAM fulfillment ratio has remained below 1.0x throughout 2025 and is projected to stay below 1.0x across all four quarters of 2026. > HBM's share of industry DRAM output is rising from 12% in 2025 to 15% in 2026, with bit growth of 103% year over year. > NAND price increases are projected to continue throughout 2026, with prices not expected to decline at any point during the year. $SNDK $SKHY $MU $DRAM $DISK $EWY

Omni 1.1 Flash is now in Google Flow, giving you even more creative control over your vision. Here’s what’s new: Introducing end frames to Omni 1.1 Flash on desktop: Direct how your video begins, and now ends, helping keep your characters and narrative consistent as you thoughtfully transition between frames. Experiment in 360p: Quickly explore new visual directions in 360p while using fewer credits. This is great for drafting on the go, right from the Google Flow mobile app. Export in 1080p and 4K: Upscale your 720p generations for a crisp, cinematic finish on final clips. Try the new updates at or download the mobile app available on Android.

🆕vLLM-Omni! Your RTX 5090 could give your Agent a quicker voice. One developer measured 47 ms to first audio. A community developer reports the median result with Qwen3-TTS 0.6B-CustomVoice. Audio starts streaming while the model is still generating later chunks. Your app can begin playback before the whole sentence is ready. For Local AI users building voice assistants, that’s a practical piece of the puzzle: getting speech out quickly on hardware you own. vLLM’s new Omni technical report explains the machinery for coordinating text, audio, image and video generation. The demo includes code, audio samples and installation instructions. Its setup requires Linux and at least 10GB of free VRAM. Link in ALT.

Spanner Omni is now generally available! The deploy-anywhere version of Spanner is here—ready to power your most demanding production workloads in your private data centers, in a multi-cloud deployment, or even for testing locally on a laptop →