Crashes, No Sound, or Screen Glitches?
Random freezes, missing sound and display glitches usually trace back to one bad driver. Find and replace yours safely.Free scan · under a minutePC Slower Than It Used to Be?
A free scan shows the junk files, broken settings and background clutter dragging Windows down - then fixes them in one click.Free scan · Windows 10 & 11No: AMD’s deal does not mean OpenAI is giving up NVIDIA GPUs. AMD and OpenAI announced a multi-year agreement for up to 6 gigawatts of AMD Instinct GPU capacity, with an initial 1-gigawatt MI450-series deployment planned for the second half of 2026. OpenAI had separately announced plans for at least 10 gigawatts of NVIDIA systems. The announcements point to a multi-supplier strategy, not an NVIDIA exit.
What AMD actually won
On October 6, 2025, AMD and OpenAI announced a definitive, multi-year, multi-generation partnership. Under the plan, OpenAI would deploy up to 6 gigawatts of AMD Instinct GPU capacity, starting with 1 gigawatt of MI450-series capacity in the second half of 2026. The companies also described a broader roadmap involving AMD Helios rack-scale systems and collaboration on hardware and software. AMD’s announcement and its filed announcement set out the public terms.
This is primarily a GPU supply and infrastructure relationship: OpenAI plans to buy and deploy AMD systems. It is not simply an investment by OpenAI in AMD. The agreement and related warrant were dated October 5, 2025, according to the product purchase agreement reference.
Planned scope and timing
- Maximum announced scope: up to 6 gigawatts of AMD Instinct GPU capacity across multiple generations.
- First milestone: 1 gigawatt using MI450-series products.
- Planned start: the second half of 2026.
- Infrastructure roadmap: AMD Helios rack-scale systems, alongside hardware and software collaboration.
These are announced plans and milestones, not evidence that the full capacity has already shipped or entered service. As of August 18, 2026, the cited official materials still identify the second half of 2026 as the planned start of the initial MI450 deployment.
Do these 3 things before closing this tab:
1Scan for outdated or missing drivers - takes under a minute2Repair Windows errors before they cause bigger problems3Fix the driver behind crashes, sound loss and screen glitches#1 Best Overall
What “6 gigawatts” means—and what it does not tell you
A gigawatt is a measure of power, not a count of processors. In this context, the figure describes the scale of computing infrastructure to be deployed. That infrastructure can include GPUs, servers, CPUs, networking, memory, cooling and power-delivery equipment.
The AMD announcement does not give a definitive total GPU count, a purchase price or enough system details to convert the capacity target into a reliable number of chips. The count would depend on the MI450 configuration, rack design, networking, power envelope and how the capacity figure is defined. It is therefore more accurate to describe the commitment in the terms AMD announced—up to 6 gigawatts—than to infer a chip total.
The AMD warrant is conditional—not 160 million shares handed over
Alongside the commercial agreement, AMD issued OpenAI a warrant to purchase up to 160 million AMD common shares. The filed warrant terms set an exercise price of $0.01 per share and an expiration date of October 5, 2030. The warrant vests in stages, with milestones tied to GPU purchases—including delivery of the initial 1 gigawatt and expansion toward 6 gigawatts—as well as AMD stock-price and performance conditions.
Rank #2
- Ideal for Gifting
- Ideal for a bookworm
- Compact for travelling
OpenAI did not receive 160 million shares immediately. Any eventual ownership percentage depends on the number of shares outstanding and on whether the relevant vesting conditions are met; describing the warrant as an automatic 10% stake would overstate what the terms establish.
Quick wins for a faster PC:
Scan for outdated or missing drivers - takes under a minuteDriver Scan →Clear out junk files and repair common Windows errorsFree Scan →AMD and NVIDIA: two announced plans, not an either-or choice
OpenAI and NVIDIA announced a separate partnership on September 22, 2025, for at least 10 gigawatts of NVIDIA systems. Their announcement said the first gigawatt was planned for the second half of 2026 and tied the systems roadmap to NVIDIA’s Vera Rubin platform. NVIDIA also announced an intention to invest up to $100 billion in OpenAI as systems are deployed; that is an announced intention, not money established as already invested. OpenAI’s announcement describes that plan.
| Announced term | AMD–OpenAI | NVIDIA–OpenAI |
|---|---|---|
| Announcement date | October 6, 2025 | September 22, 2025 |
| Capacity target | Up to 6 GW of AMD Instinct GPU capacity | At least 10 GW of NVIDIA systems |
| Initial milestone | 1 GW | 1 GW |
| Planned initial timing | Second half of 2026 | Second half of 2026 |
| Hardware identified | Instinct MI450 series initially, with future generations | NVIDIA systems; announcement tied the roadmap to Vera Rubin |
| Equity component | Warrant for up to 160 million AMD shares, subject to vesting conditions | NVIDIA intention to invest up to $100 billion in OpenAI as systems are deployed |
| Exclusivity stated in cited public announcement? | No | No |
| Completion established by the cited announcement? | No; initial rollout was scheduled for the second half of 2026 | No; initial rollout was scheduled for the second half of 2026 |
The public AMD announcement calls AMD a “core strategic compute partner,” but does not say it is OpenAI’s sole GPU supplier. The separate NVIDIA plan is direct evidence that the two relationships were announced as concurrent plans. The announcements do not establish whether the capacity will serve different sites, workloads or phases, nor whether the targets overlap operationally.
Why OpenAI may want both suppliers
At this scale, using more than one accelerator supplier can be a practical way to expand available capacity and reduce dependence on a single hardware roadmap. It can also strengthen procurement flexibility and give OpenAI options to match different systems to different training or inference workloads. Those are strategic interpretations of the arrangement, not a published breakdown of how OpenAI will assign workloads.
Multiple vendors also add complexity. AMD’s ROCm software ecosystem differs from NVIDIA’s; adapting kernels, libraries, compilers and fleet operations takes engineering effort. OpenAI would need to manage distinct software and support environments, while AMD would need to demonstrate that its systems perform reliably at production scale. Power efficiency and total system cost may matter, but the announced materials do not provide comparable prices or benchmark results.
What the agreement could mean financially
AMD said the partnership could generate tens of billions of dollars in revenue. That is a forward-looking company expectation, not a disclosed fixed contract value, guaranteed revenue or annual run rate. The eventual result would depend on actual purchases and deliveries, deployment timing, product pricing and whether the full capacity target is reached. The filed announcement does not support calling the AMD deal a $100 billion contract.
Rank #4
For investors, the warrant adds a potential equity component, but its value is conditional on vesting and share-price milestones. Neither the revenue outlook nor the warrant makes the outcome certain: manufacturing capacity, data-center readiness, financing and execution all affect what is ultimately delivered.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.What has not been established yet
The announcements establish a significant commercial plan, not proof of completed deployment. The cited official materials do not establish:
- That the initial AMD gigawatt has shipped or is running production workloads.
- The final MI450 specifications, independent performance results or exact number of GPUs.
- The purchase price, exact OpenAI purchase obligations or deployment locations.
- That all 6 gigawatts will be deployed, or that NVIDIA’s separate plan has been canceled.
A signed agreement can still be delayed, modified or only partly implemented. The clearest evidence of execution will be confirmed shipments, systems running OpenAI workloads, disclosed deployment milestones and financial reporting that attributes revenue to the program. MI450 benchmarks, ROCm production results, warrant-vesting updates and confirmation that NVIDIA’s plan remains on schedule would further show how the competing platforms fare in practice.
Recommended Free Tools
Best Value
- It can be a gift option
- Comes with secure packaging
- Helpful in various ways
Who stands to gain—and what could go wrong
OpenAI
OpenAI gains another potential source of accelerator capacity and more flexibility in supplier negotiations. The trade-off is the work of bringing another hardware and software ecosystem into production, with added operational demands and the risk that a product or rollout misses expectations.
AMD
The agreement gives AMD a high-profile customer and a chance to validate Instinct systems at unusual scale. That could improve its credibility with other AI infrastructure buyers. The opportunity depends on delivering MI450 systems on schedule and making the complete platform—including software, networking and support—work reliably. Aggressive pricing, supply limits or delays could reduce the commercial benefit.
NVIDIA
AMD’s selection increases competitive pressure, but it does not amount to a loss of OpenAI as a customer: OpenAI separately announced its NVIDIA systems plan. The more meaningful competitive test is whether AMD can deliver capacity that meets OpenAI’s performance, reliability and cost requirements at scale.
How to judge whether AMD has truly won
For readers tracking the agreement, distinguish each step rather than treating an announcement as a completed deployment:
Free tools Windows power users keep installed
One-click scans. No signup required.
Quick Recap
- Commitment: What purchase obligations and conditions are specified in filings and subsequent updates?
- Delivery: Has AMD confirmed shipment and installation of the initial 1 gigawatt?
- Production use: Has OpenAI confirmed that its workloads run on the systems?
- Performance and economics: Are comparable workload results, utilization, reliability and total system costs available?
- Scale: Are later purchase milestones being met, and is OpenAI expanding beyond the initial deployment?
- Financial follow-through: Does AMD report attributable sales, and do filings show warrant milestones vesting?
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
