Reports
AI-generated structured vendor updates
Samsung GAIA AI PC Chip Samples with Memory-Centric NPU, Targeting 50 TOPS
Samsung launches GAIA AI PC processor with 4nm process and memory-centric NPU, integrating LPDDR5X controller with NPU for near-memory computing, achieving 40% energy efficiency improvement and 50 TOPS. Certified for Microsoft Copilot+ PC, Lenovo to adopt in Q4 2026.
Towards Feature Complete Triton Support in JAX-Triton â ROCm Blogs
...
NVIDIA Vera CPU: Max Single-Threaded Performance at Scale for Agentic AI
NVIDIA launches Vera CPU, a max single-threaded CPU at scale for agentic AI. With Olympus cores delivering 1.8x sustained per-core performance over x86, 1.2TB/s LPDDR5X bandwidth, and 3.4TB/s core-to-core bandwidth, Vera integrates into NVIDIA's unified AI factory architecture, aiming to lock users into its ecosystem.
AI Innovators Adopt NVIDIA Vera — Why Max Single-Threaded CPU at Scale Matters
...
OpenAI Winds Down Fine-Tuning API: A Strategic Shift in AI Customization Landscape
OpenAI plans to phase out its fine-tuning API by 2027, stopping new task creation but allowing inference on existing models. This forces startups relying on fine-tuning for differentiation to migrate to open-source models or RAG, reshaping the AI customization ecosystem.
Qualcomm Enters AI Inference with Dragonfly C1000 CPU and HBC Near-Memory Compute
Qualcomm unveils Dragonfly roadmap with Oryon-based C1000 CPU and AI300 inference accelerator featuring HBC near-memory compute. Meta and Microsoft are early adopters. The strategy targets AI inference TCO reduction and memory wall breakthrough, bypassing Nvidia's training dominance.
Etched Unveils Sohu Transformer ASIC: Claims 20x H100 Inference Throughput, Challenging NVIDIA's Grip
AI chip startup Etched emerges from stealth with Sohu, a Transformer-specific ASIC on TSMC N4P with 144GB HBM3E. By hardwiring attention mechanisms, it claims 20x throughput and 140x price-performance vs. H100 on Llama 70B. With $800M total funding and first racks shipping this summer, it directly challenges NVIDIA's inference dominance.
Making private MCP servers reachable without making them public | OpenAI Developers
...
Qualcomm Enters AI Datacenter with Dragonfly ARM CPU, Meta Signs Multi-Generation Deal
Qualcomm unveils Dragonfly C1000 ARM-based datacenter CPU, AI300 accelerator, and interconnect. Meta commits to multi-generation CPU supply, Microsoft Azure to deploy HBC chips. Qualcomm targets $15B+ datacenter revenue by FY2029, acquires Modular for software stack.
Huawei Pushes Token-Based Billing at MWC Shanghai 2026: Shifting Carrier Monetization from Bytes to AI Inference Value
At MWC Shanghai 2026, Huawei urged carriers to shift from byte-based to token-based billing for AI workloads, showcasing a 372% token throughput improvement in long-sequence inference via its AI Inference Acceleration Solution. It also highlighted the Upper-6 GHz band as critical for AI wearables requiring 20 Mbps uplink, aiming to reposition 5G-A networks as AI compute delivery infrastructure.
Qualcomm HBC Gen 1 Stacks LPDDR to 133 TB/s, Challenging HBM Dominance
Qualcomm announces HBC Gen 1, a 3D-stacked LPDDR memory with integrated compute die, achieving 133 TB/s bandwidth and 6x energy efficiency over HBM. Aimed at replacing HBM in AI accelerators, shipping with AI250 in mid-2027, but supply chain and feasibility remain uncertain.
Anthropic Alleges Largest AI Distillation Attack by Alibaba-Linked Operators, Exposing API Security Gaps
Anthropic alerted U.S. senators that Alibaba-linked operators conducted the largest known distillation attack, generating 28.8 million model exchanges via 25,000 fraudulent accounts to harvest Claude's frontier capabilities. The incident exposes a critical vulnerability in AI API security, forcing a rethinking of inference endpoint protection and usage monitoring.
Huawei Unveils AI-Centric Network with Token Monetization, UCM Caching Breaks Long-Context Barriers
At MWC Shanghai 2026, Huawei unveiled an AI-native network architecture integrating service, network, and compute, shifting from traffic-centric to intelligence-centric operations. The Unified Cache Manager (UCM) extends KV cache to petabyte-scale external storage, achieving 372% token throughput gains on GLM-5.1 at 128K sequence lengths. Token monetization frameworks and agentic operations enable carriers to charge for AI inference capacity and personalize services.
Google Cloud Multi-Agent Architecture Shifts Control from Human to Autonomous Verification
Google Cloud introduces agent-scale data management with multi-agent verification to reduce human oversight. Deploys six Gemini agents with Nokia for autonomous network operations. Amazon plans to commercialize Trainium chips, intensifying AI hardware competition against Google TPU and Nvidia GPU.
Qualcomm Dragonfly: 250-core CPU, HBC memory, UALink interconnects target AI inference TCO
Qualcomm unveils full data center portfolio: Dragonfly C1000 250-core Oryon CPU (>5GHz, PCIe Gen7, CXL), HBC near-memory compute (133TB/s Gen1, 18x-54x effective BW), AI300 inference accelerator (UALink/ESUN scale-up), and 800G/1.6T connectivity. Multi-year Meta CPU deal. Commercial sampling 2027-2028. Targets inference TCO with tokens-per-watt leadership.
OpenAI and Broadcom Unveil Jalapeno Inference ASIC, Reshaping AI Hardware Landscape
OpenAI, in collaboration with Broadcom, has developed Jalapeno, a custom LLM inference accelerator. The chip uses a multi-chip module with HBM3E memory and achieved tape-out in just nine months. Designed for OpenAI's model stack, it aims to reduce inference costs and dependency on NVIDIA GPUs, with initial deployment planned for late 2026.
Cisco Live US & InfoComm 2026 : la collaboration entre dans l’ère agentique
...
Mandiant Reveals Cisco SD-WAN Manager Zero-Day: Control Plane Becomes Prime Target
Mandiant identified a zero-day (CVE-2026-20245) in Cisco Catalyst SD-WAN Manager exploited via malicious CSV upload to escalate to root. The intrusion involved rogue peering, credential manipulation, and anti-forensic cleanup. This highlights SD-WAN centralized control planes as a new attack surface for advanced threats.
NVIDIA and AWS Default GPU Vector Search with cuVS, G7 Instances Deliver 4.6x Inference
NVIDIA and AWS collaborate to embed cuVS as default GPU-accelerated vector search in OpenSearch Serverless, delivering 10x faster indexing at 1/4 cost. New EC2 G7 instances with RTX PRO 4500 Blackwell GPUs achieve up to 4.6x inference performance. AWS achieves GB300 Exemplar Cloud status for training.
China's LineShine Tops TOP500: CPU-Only 2.2 ExaFLOPS with ARMv9 and HBM Memory
LineShine supercomputer achieves 2.198 ExaFLOPS FP64 sustained using 13.79 million ARMv9 cores across 20,480 nodes, making it the first system to exceed 2 ExaFLOPS without GPUs. Each node has dual LX2 CPUs (304 cores) with 32GB HBM, demonstrating a CPU+HBM architecture breakthrough for HPC.