Blog
29.May.2025
🔍 AI Tech Highlights from COMPUTEX 2025
COMPUTEX 2025 embraced the theme "AI Next," showcasing how artificial intelligence is reshaping industries from cloud to edge. In this edition of C&T Tech Hunt, we’re spotlighting some of the most compelling innovations we discovered on the show floor from AI-powered workstations and voice-controlled robotics to next-gen datacenter platforms and sustainable computing breakthroughs.
1️⃣ Local AI Agents Get a Boost with Intel® Xeon® 6 & Intel AMX
Intel introduced its latest Xeon 6 series processors, engineered to power high-performance, AI-centric infrastructures.
💡 What’s exciting?
The new Xeon 6 platforms support Intel Advanced Matrix Extensions (AMX), a specialized hardware instruction set for accelerating matrix operations perfect for running AI inference workloads natively on CPU.
🔍 Use Case Highlights:
- Classical Machine Learning Models: From large-scale data analytics to medical diagnostics and virtual desktop infrastructure (VDI), the Xeon 6 shines in compute-heavy tasks.
- Small Generative AI Models (<20B parameters): Think on-premise chatbots, summarization tools, and search embedding engines all without needing discrete GPUs.
- Large GenAI Hosting: While larger models still depend on AI accelerators like NVIDIA GPUs, Xeon 6 serves as a powerful orchestration host CPU for distributed systems.
2️⃣ AI Playground with Intel® Arc™ Graphics: Create, Enhance, and Explore with Local AI
🧠 What is AI Playground?
It’s a suite of hands-on demos that showcase real-time, local AI applications—from text-to-image generation to intelligent document editing all running without the need for cloud access.
🛠️ Key Highlights:
- Create with AI: Generate art, posters, or album covers in seconds by turning simple text prompts into vivid, AI-generated images.
- Enhance with AI: Use AI upscaling and editing tools to transform photos—retouching, expanding image boundaries, and sharpening resolution.
- Answer with AI: Summarize documents, generate intelligent responses, and refine content using uploaded text files and AI-assisted writing tools.
- Customize AI Models: Developers can swap in alternative AI models, taking full advantage of Intel’s local acceleration for more advanced experimentation.
The AI Playground proves that AI innovation isn’t just for the data center with the right hardware, every PC can become an AI workstation, enabling productivity, creativity, and experimentation right at your fingertips.
3️⃣ Voice-Controlled Robot Arm That Sees and Understands
🎙️ Speak to Act:
This robot interprets voice commands and performs physical movements like picking and placing objects. But it doesn’t stop there it also uses AI-based vision to recognize color and shape, ensuring accurate execution of tasks.
🤖 Applications:
- Smart logistics
- Assistive robotics
- Factory automation
It’s like combining Natural Language Processing + Computer Vision + Robotics into one seamless interface making human-machine collaboration even more intuitive.
4️⃣ NVIDIA HGX B300 Platform: A Glimpse Into the Future of AI Data Centers
NVIDIA showcased its powerful HGX B300 platform, built to handle the most demanding AI training and inference workloads at scale especially for Large Language Models (LLMs) and Generative AI.
🧠 What is HGX B300?
It’s the latest AI supercomputing platform combining:
- 16x NVIDIA Blackwell GPUs
- 2x NVLink Switch Chips for ultra-fast inter-GPU communication
- 105 PFLOPS FP4 performance for dense AI inference
- 2.3 TB of HBM3e memory
- Support for air-cooled and liquid-cooled configurations
- 8x ConnectX-8 SuperNICs for high-speed networking
- Foundation model training (LLMs, VLMs)
- Hyperscale inference workloads
- Scientific research and simulations
- Data center consolidation with maximum compute density
Despite its power, the B300 focuses on performance-per-watt, aligning with the growing push for sustainable AI computing.
5️⃣ Sustainable AI Computing with GIGABYTE B683 Blade Servers
🧊 Cooling Innovation:
These blades feature direct-to-chip liquid cooling (DLC) for both CPU and memory, dramatically improving heat dissipation.
🏗️ Key Stats:
- Fits 20 CPUs in a 6U chassis
- High-density compute with modular cooling
- Optimized for AI workloads at scale
With global data centers consuming more power than ever, efficient cooling and green compute will be the foundation of sustainable AI infrastructure.
6️⃣ Fluorinated Immersion Cooling Liquid "DAISAVE": A Cooler Future for Data Centers
🌊 What is DAISAVE?
DAISAVE is Daikin’s proprietary brand of fluorinated dielectric fluids designed specifically for single-phase immersion cooling. These liquids are non-conductive, thermally efficient, and feature low Global Warming Potential (GWP)—offering both performance and environmental sustainability.
🔥 Use Case Highlights:
- Enhanced Cooling Efficiency: By removing the need for traditional fans, heat sinks, or chillers, DAISAVE delivers direct-to-liquid thermal exchange that significantly reduces energy consumption.
- Corrosion Resistance: The oxygen- and moisture-free environment protects electronic components from rust and corrosion, increasing reliability and lifespan.
- Silent Operation: With fewer moving parts, immersion systems are quieter, making them ideal for edge data centers in noise-sensitive environments.
- Space Optimization: Higher cooling efficiency allows for denser server packing, saving valuable real estate in hyperscale deployments.
Final Thoughts
This year’s COMPUTEX made one thing clear: AI is no longer future tech it’s infrastructure. Whether on the factory floor, in creative studios, or inside massive data centers, the next wave of AI is about bringing intelligence closer to where decisions happen.
Which trend are you most excited about?
Let’s keep the conversation going. Drop us a message or connect with our team to explore how edge AI and industrial computing are shaping the next era of smart systems.
