Slicast
Monday, September 28, 2026
DarkSubscribe
Slicast
AI Infrastructure · News & Analysis
Home › Chips & Hardware › Report
Chips & Hardware · Report

Broadcom and OpenAI unveil Jalapeño custom AI chip targeting 50% cheaper inference and reduced NVIDIA dependency.

OpenAI's custom chip validates in-house accelerator strategy; 50% cost reduction threatens NVIDIA inference market.
Trade pressSlicast · June 27, 2026 at 14:00 UTC · US · Source: Barchart.com
importance 78
Read the original
Companies in this story
NvidiaCapex history →
BroadcomCapex history →
OpenAI
In-depth analysis
CommentaryNvidia vs Alibaba Zhenwu V900: Memory Claims and China's AI Chip Independence, September 2026CommentaryWhy Nvidia's AI Energy Management Alliance Signals That Grid Capacity, Not Silicon, Now Constrains AI Scale
Related reports
Chips & HardwareOpenAI unveiled custom AI inference chip ('Jalapeño') targeting 50% cost reduction versus Nvidia GPUs; GPU rental prices fall 31%2026-06-28Chips & HardwareNVIDIA chip prices forecast to decline as competition intensifies from AMD, custom silicon, and inference alternatives.2026-06-28Chips & HardwareBroadcom unveils Jericho3-AI Ethernet switch with 800G ports, enabling AI cluster interconnect at new bandwidth density.2026-06-28Chips & HardwareOpenAI unveils Jalapeño, its first custom-designed AI inference chip2026-06-28Chips & HardwareNVIDIA announces Vera Rubin GPU platform, a new high-performance accelerator for advanced AI compute workloads.2026-06-28
Broadcom and OpenAI unveil Jalapeño custom AI… · Slicast