Nvidia GPUs Can Report Individual GDDR6 and GDDR6X Memory Chip Temperatures, Enthusiasts Discover

• July 19, 2026

Recent developments within the Nvidia graphics card community have revealed an advanced temperature monitoring feature previously undocumented by the manufacturer. After enthusiasts managed to restore the ability to measure the Hotspot temperature on Nvidia’s Blackwell-generation GPUs, another team of researchers from Brazil, known as the Paulo Gomes Team, identified a new hidden capability. This functionality enables Nvidia GPUs to individually report the temperature of each GDDR6 and GDDR6X memory chip, a more granular level of detail than what typical monitoring tools currently display.

Detailed Memory Chip Temperature Reporting Now Accessible

Standard monitoring utilities for Nvidia graphics cards have traditionally provided only a single aggregate value representing the maximum temperature across all memory chips. However, the newly uncovered feature allows for the precise temperature data of each individual memory chip to be accessed, offering enhanced insights into the operating conditions of high-speed GDDR6 and GDDR6X modules.

This discovery adds an important dimension to thermal management and reliability tracking for high-performance graphics hardware. Each memory chip operates independently and can experience varying levels of thermal stress depending on load distribution, air flow, and other system factors. Having access to per-chip temperature data could assist enthusiasts, overclockers, and system integrators in fine-tuning cooling solutions or diagnosing thermal-related performance issues with greater precision.

The finding follows earlier breakthroughs where users reinstated the measurement of the GPU Hotspot temperature— a critical indicator of the hottest point on the GPU die— on Nvidia’s latest Blackwell GPU family. The Blackwell architecture represents the current generation of Nvidia graphics processors, and expanding sensor access within this hardware paves the way for more comprehensive hardware monitoring capabilities.

The Paulo Gomes Team’s efforts involved deep analysis of GPU firmware and sensor data protocols to enable this enhanced telemetry. Their work reveals that Nvidia GPUs incorporate internal sensors that monitor not just core temperature but also detailed memory module thermal states, though these have remained largely inaccessible to end users until now.

While Nvidia’s official monitoring tools and third-party applications have yet to natively implement support for displaying each memory chip’s temperature separately, these findings provide a realistic pathway for such features to be introduced. This could improve hardware diagnostics during stress testing, gaming, or professional workloads where memory temperature can influence stability and longevity.

For now, the discovery remains within the enthusiast community, with further development required to integrate this capability into user-friendly monitoring software. Nonetheless, it highlights the evolving landscape of GPU telemetry and the increasing level of detail accessible to users interested in maximizing hardware performance and reliability.

Enthusiasts uncover hidden capability in Nvidia GPUs to monitor temperatures for each GDDR6 and GDDR6X memory chip separately.