AI BriefingQualcomm Launches Cloud AI 100 Ultra Inference Card with 128GB Memory and Up to 870 TOPS
·2026.11.05 04:55
Key point
The new PCIe card features 64 AI cores, 576 MB on-die SRAM, and 548 GB/s DRAM bandwidth for generative AI workloads.
Details
- Qualcomm releases the Cloud AI 100 Ultra, a 150W PCIe AI inference card designed for scaling generative AI and LLM workloads.
- The card delivers up to 870 TOPS (INT8) and 288 TFLOPS (FP16) using 64 AI cores and 576 MB of on-die SRAM.
- Equipped with 128 GB of LPDDR4x DRAM at 548 GB/s bandwidth, it connects via PCIe Gen4 x16.
- The accompanying AI Inference Suite provides OpenAI-compatible APIs, RAG support, and Kubernetes deployment options for enterprise agents.
- An AI 80 Ultra variant is also available with 618 TOPS and 222 TFLOPS performance.
This summary was generated automatically by AI. Check the original for the author's claims and context. Copyright belongs to the original author.
Our guide explains how the AI works. Report summary errors, attribution issues, or removal requests via Contact.
Related
Training 1,000 Weights in Gemma 3 270M Unlocks Hidden Sorting and JSON SkillsBaseten Launches Grid, Sandboxes, and Serverless Loops