Reports
AI-generated structured vendor updates
SambaNova完成11亿美元融资估值110亿美元：推理芯片新格局确立
...
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.
Qualcomm Acquires Modular for $3.9B, Open-Sources Mojo to Break CUDA Lock-In
Qualcomm acquires Modular for $3.9B in stock and open-sources Mojo, a Python-compatible systems language. Mojo targets CUDA dependency, aiming to provide a high-performance alternative for AI developers. This move strengthens Qualcomm's AI inference chip software stack and edge AI competitiveness.
OpenAI and Broadcom unveil Jalapeño inference ASIC to bypass NVIDIA GPU dependency
OpenAI and Broadcom launch Jalapeño, a custom ASIC for LLM inference, achieving tape-out in 9 months. OpenAI designs architecture, Broadcom provides networking, Celestica handles integration. Planned for large-scale deployment by end-2026 with gigawatt-scale datacenters, aiming to cut inference costs and reduce NVIDIA dependency.
AWS Launches Inferentia2 Chip for Generative AI Infrastructure Optimization
AWS launched second-gen Inferentia2 AI inference chip, designed for Transformer models with 4x performance boost and support for 175B parameter models. Integrated into EC2 Inf2 instances with UltraClusters architecture for large-scale deployment, offering 40% better cost-performance and 50% lower power consumption than GPU instances.
OpenAI and Broadcom Launch Jalapeño ASIC for LLM Inference, 9-Month Tapeout
OpenAI and Broadcom unveil Jalapeño, a custom ASIC for LLM inference, achieving tapeout in 9 months. The chip reduces data movement and claims superior performance per watt. Deployment planned by end of 2026, marking OpenAI's shift to integrated hardware-software infrastructure.
NVIDIA Acquires Groq LPU: Inference Architecture Shift from HBM to On-Chip SRAM
NVIDIA signs ~$20B licensing deal with Groq for LPU tech, featuring 230MB on-chip SRAM at 80TB/s bandwidth. This targets Transformer inference decode, replacing HBM bottlenecks with ultra-low latency on-chip storage, potentially reshaping the AI inference chip landscape.