VQV Signal
According to a post on X by @zephyr_z9, NVIDIA cut NVLink bandwidth per GPU from 1.8 TB/s to 1.5 TB/s. The post follows a report that NVIDIA lowered HBM4 pin speed from 11.7 to 9.3 Gbps. By the author's numbers, AMD would lead on both HBM bandwidth and scale-up bandwidth. NVIDIA has not commented, and the post itself says both claims should be treated as reports until confirmed.
According to a post on X by @zephyr_z9, NVIDIA cut NVLink bandwidth per GPU from 1.8 TB/s to 1.5 TB/s. The post follows a report that NVIDIA lowered HBM4 pin speed from 11.7 to 9.3 Gbps. By the author's numbers, AMD would lead on both HBM bandwidth and scale-up bandwidth. NVIDIA has not commented, and the post itself says both claims should be treated as reports until confirmed.
AI-assisted summary based on the listed source.
Following the report that NVIDIA lowered HBM4 pin speed from 11.7 to 9.3 Gbps, @zephyr_z9 says it also cut NVLink bandwidth per GPU from 1.8 TB/s to 1.5 TB/s, which by his numbers leaves AMD ahead on both HBM bandwidth and scale-up bandwidth. NVIDIA hasn't commented, so treat both as reports until confirmed; they...
If the figures hold, the changes would affect NVIDIA, AMD and memory-chip makers, the companies the post names as most exposed. Buyers comparing AI accelerators would need to weigh these specs against the vendors' official datasheets. Watch for an NVIDIA response or confirmation of the HBM4 and NVLink specifications.
Hardware and robotics watchers may want to track whether this becomes a product, benchmark, or deployment signal.
VQV organizes public signals from inspectable sources. It does not independently verify the underlying report.
Signal Strength reflects source quality, relevance, freshness and evidence. Public Interest helps organize discovery; it is not proof of truth.
VQV surfaced this signal because it is recent, relevant to AI Chips, connected to Scobleizer News Pulse (X).
No login, cookies, social SDKs, or automatic posting.
