Get the newsletter
Nvidia · Chip · Filed Apr 7, 2026 · USPTO published Aug 20, 2026 · Aug 20 release

Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters

Your GPU wastes power running low-priority background tasks at full tilt. Nvidia wants chips that watch what kind of work is coming and dial power up or down accordingly, before the task even starts.

Two identical processor chips housed together, each operating at 2.8 gigahertz with 72 threads and 250 watts of power. Drawing from patent filing US 2026/0244254 A1.
Two identical processor chips housed together, each operating at 2.8 gigahertz with 72 threads and 250 watts of power.All 45 drawings ↓
Publication number
US 2026/0244254 A1(US20260244254A1)
Applicant
NVIDIA Corporation
Filed · Published
Apr 7, 2026 · Aug 20, 2026
Inventors
Sreedhar Narayanaswamy
Status when we published
Waiting for an examiner
Document
20 claims
US classification
713/330
Parent application
Continuation of 18/425,618 (filed Jan 29, 2024)(18425618)

What Nvidia's thread-priority power system actually does

Chips today treat almost every task the same for power: they ramp up, do the work, and ramp down, without much sense of whether the task actually needed that energy. Nvidia wants to change that by teaching chips to check the importance of incoming work before deciding how much electricity to use.

Here's the idea: every program you run creates small units of work called threads. Some threads are time-sensitive and important (say, rendering the next frame of a game, or running an AI inference request someone is waiting on). Others are background housekeeping that can wait. Nvidia's patent describes a system where the chip gets a heads-up about what mix of high-priority and low-priority threads is coming, then sets its power level to match.

The practical upside is that a chip handling mostly low-priority work doesn't need to burn full power. And when important work arrives, it gets the headroom it needs without delay. That balance matters most in data centers running mixed workloads, but the principle applies anywhere Nvidia chips operate.

From the filing · CLAIM 1
obtain information indicating an expected proportion of thread priorities of threads to be performed by a first one or more processors; determine an amount of power to allocate to the first one or more processors based, at least in part, on the information indicating the expected proportion of thread priorities; …

Translation: The chip checks how important the upcoming tasks are so it knows how much electricity to send.

How the chip reads task priority and adjusts its own power draw

The core mechanism is an API (a software interface that lets programs talk to hardware) through which software can tell the processor what proportion of upcoming threads are high-priority versus low-priority. The chip then uses that proportion to calculate how much power to allocate and adjusts the actual voltage and frequency delivered to match.

The patent's key phrase is "expected proportion of thread priorities." Rather than reacting after the fact, the chip gets a forward-looking signal. If 80% of the next batch of threads is low-priority background work, the chip doesn't need to sustain its maximum power state. If a burst of high-priority threads is incoming, it can pre-position itself to deliver more power without a lag.

  • Obtain: The processor reads information about the priority mix of threads about to be scheduled.
  • Determine: It calculates a target power level suited to that mix.
  • Adjust: It changes the actual power delivered to match that target before or as the work runs.

The patent frames this as an API-level capability, meaning software developers or runtime environments (like AI inference servers or game engines) could explicitly communicate priority information to the hardware, giving the chip better data than it could infer on its own.

From the filing · THE ABSTRACT
Apparatuses, systems, and techniques to allocate power to one or more processors. In at least one embodiment, processors or computing systems perform an API to allocate power to one or more processors based, at least in part, on indications of priority of one or more threads to be performed by the one or more processors.

Translation: Software tools help route extra electricity to the chips when high priority tasks arrive.

What this means for AI workloads and your GPU's efficiency

Power management in chips is a constant tug of war between performance and efficiency. Data centers running Nvidia GPUs for AI training or inference face real costs when chips idle at high power states during low-priority work, or when they're slow to ramp up for high-priority requests. A system that pre-reads the workload mix and adjusts power proactively could reduce energy waste on the low end and cut latency on the high end, both of which translate to money and user-facing responsiveness.

For everyday users, the most noticeable version of this kind of system would be a GPU that doesn't drain laptop battery during background tasks but still delivers full performance the moment you need it. The filing sits within a broader trend of chip-level intelligence around power, one well tracked through new Big Tech patents covering silicon efficiency across Nvidia, AMD, Intel, and Apple.

Nvidia's 28th filing we've tracked since July in our AI chip wars watchlist follows one cutting AI response time and one on two-stage video compression.

Source: US 2026/0244254 A1 at the USPTO (PDF). Full patent text and figures are in the official publication.

Get our take in your Top Stories

Liked this breakdown? Add Patentlyze as a preferred source on Google, and our plain-English take shows up more often in your Top Stories the next time Nvidia patent news breaks.

There are more where this came from

We read the new patents from Big Tech, carmakers and more and send you the ones worth knowing. Plain English, free, every week.

The drawings · 45 sheets

45 drawing sheets from US 2026/0244254 A1 · click any drawing to enlarge

Dual-chip processor hardware layout, drawing sheet 1 from US 2026/0244254 A1
Sheet 1 Dual-chip processor hardware layout
Superchip power allocation schematic, drawing sheet 2 from US 2026/0244254 A1
Sheet 2 Superchip power allocation schematic
Firmware power adjustment workflow, drawing sheet 3 from US 2026/0244254 A1
Sheet 3 Firmware power adjustment workflow
Power estimation and capping process, drawing sheet 4 from US 2026/0244254 A1
Sheet 4 Power estimation and capping process
Controller hierarchy and power distribution, drawing sheet 5 from US 2026/0244254 A1
Sheet 5 Controller hierarchy and power distribution
Multi-controller power routing layout, drawing sheet 6 from US 2026/0244254 A1
Sheet 6 Multi-controller power routing layout
Operating system thread priority mapping, drawing sheet 7 from US 2026/0244254 A1
Sheet 7 Operating system thread priority mapping
Data center hardware resource hierarchy, drawing sheet 8 from US 2026/0244254 A1
Sheet 8 Data center hardware resource hierarchy
Show the other 37 sheets
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 9
Sheet 9
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 10
Sheet 10
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 11
Sheet 11
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 12
Sheet 12
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 13
Sheet 13
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 14
Sheet 14
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 15
Sheet 15
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 16
Sheet 16
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 17
Sheet 17
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 18
Sheet 18
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 19
Sheet 19
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 20
Sheet 20
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 21
Sheet 21
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 22
Sheet 22
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 23
Sheet 23
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 24
Sheet 24
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 25
Sheet 25
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 26
Sheet 26
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 27
Sheet 27
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 28
Sheet 28
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 29
Sheet 29
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 30
Sheet 30
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 31
Sheet 31
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 32
Sheet 32
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 33
Sheet 33
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 34
Sheet 34
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 35
Sheet 35
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 36
Sheet 36
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 37
Sheet 37
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 38
Sheet 38
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 39
Sheet 39
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 40
Sheet 40
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 41
Sheet 41
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 42
Sheet 42
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 43
Sheet 43
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 44
Sheet 44
Nvidia Patents a Way to Feed Its Chips More Power When the Work Actually Matters patent drawing, sheet 45
Sheet 45
Patent filing page

Watchlists this filing belongs to

All watchlists →