Recommended Free Tools
iTechGuides is reader-supported. When you buy through links on our site, we may earn an affiliate commission. As an Amazon Associate I earn from qualifying purchases. Learn more
Blackwell was already a shipping, revenue-producing platform in calendar Q1 2025—not a paper launch. NVIDIA had begun shipping production Blackwell systems in its fiscal fourth quarter, which ended January 26, 2025. In results released February 26, the company reported $11.0 billion in Blackwell architecture revenue within $39.331 billion of quarterly revenue.
The name covers several different products. B200 is the data-center GPU; GB200 combines two B200 GPUs with a Grace CPU; and GB200 NVL72 is a liquid-cooled rack containing 72 Blackwell GPUs. The same architecture also reached consumer PCs through the GeForce RTX 50 Series.
What “Q1 2025” means here
This article uses calendar Q1 2025: January through March. NVIDIA’s February 26 financial release used a different period—fiscal Q4 FY2025, ended January 26, 2025. Treating that release as calendar Q1 revenue would mix two calendars.
Blackwell’s status during the quarter
Production had started before January
NVIDIA’s SEC filing states: “We began shipping production systems of the Blackwell architecture in the fourth quarter of fiscal year 2025.” Because that fiscal quarter ended January 26, production shipments were already under way at the start of calendar Q1.
#1 Best Overall
- PLEASE NOTE: Exporting an NVIDIA RTX Pro 6000 GPU outside the US requires strict adherence to the U.S. Export Administration Regulations (EAR) and issuance of an export license from the Bureau of Industry and Security (BIS). Compliance and Know Your Customer (KYC) screening may be required as a condition of order acceptance. [NVIDIA Blackwell Streaming Multiprocessor] The new SM features increased processing throughput, and new neural shaders that integrate neural networks inside of programmable shaders | DLSS 4: Multi Frame Generation ensures ultra-smooth frame pacing for lifelike simulations.
- [Double-Flow-Through Design] The RTX PRO 6000 Blackwell features a double-flow-through cooling design, optimizing efficiency and airflow to sustain peak performance under 600W power loads. | [5th Gen Tensor Cores] Deliver up to 3X the performance of the previous generation and support for FP4 precision for faster AI model processing times with reduced memory usage, enabling local fine-tuning of LLMs and generative AI | [4th Gen Ray Tracing Cores] Double the ray-triangle intersection rate of the previous generation to create photoreal, physically accurate scenes and immersive 3D designs with RTX Mega Geometry, which enables up to 100X more ray-traced triangles.
- [PCIe Gen 5] Support for PCIe Gen 5 provides double the bandwidth of PCIe Gen 4, improving data-transfer speeds from CPU memory and unlocking faster performance for data-intensive tasks like AI, data science, and 3D modeling. | [GDDR7 Memory] With 96 GB of GPU memory and 1.8 TB ps bandwidth, it can tackle massive 3D and AI projects, fine-tune AI models locally, explore large-scale VR environments, and drive larger multi-app workflows.
- [DisplayPort 2.1] Achieve unparalleled visual clarity and performance, driving high resolution displays at up to 8K at 240 Hz and 16K at 60 Hz. Increased bandwidth enables seamless multi-monitor setups while HDR and higher color depth support ensures superior color accuracy for precision work, such as video editing, 3D design, and live broadcasting.
- [Universal MIG] Divide a single RTX PRO 6000 Blackwell into multiple isolated instances, each with dedicated resources, allowing for concurrent execution of multiple workloads, optimized GPU utilization, and secure isolation of different applications or users. [WARRANTY] 3 YR Manufacturer's Warranty. Bulk OEM Packaging. Retail Packaging is NOT included.
Blackwell generated billions in reported revenue
On February 26, NVIDIA reported $11.0 billion of Blackwell architecture revenue for its fiscal Q4 FY2025 quarter. The company’s total quarterly revenue was $39.331 billion. These are NVIDIA-reported financial figures for the quarter ended January 26, not a complete January–March unit-shipment tally.
“Demand for Blackwell is amazing as reasoning AI adds another scaling law — increasing compute for training makes models smarter and increasing compute for long thinking makes the answer smarter.”
— Jensen Huang, NVIDIA founder and CEO, February 26, 2025
What’s actually slowing this PC down?
Pick the symptom - the matching free tool is one click away.
Special offer. See more information about Outbyte and uninstall instructions. Please review EULA and Privacy policy.
Cloud access became practical
On February 4, NVIDIA said CoreWeave had made GB200 NVL72 instances generally available. NVIDIA described CoreWeave as the first cloud service provider to make the Blackwell platform generally available. For organizations that could not install liquid-cooled racks, renting an instance was therefore an available access route during the quarter.
Which products are actually “Blackwell”?
| Product | What it is | Where it fits |
|---|---|---|
| B200 | One NVIDIA Blackwell Tensor Core GPU | Building block for data-center servers and systems |
| GB200 Grace Blackwell Superchip | Two B200 GPUs connected to one Grace CPU | Superchip used in high-end AI systems |
| GB200 NVL72 | 36 GB200 superchips, totaling 72 Blackwell GPUs and 36 Grace CPUs | Liquid-cooled rack-scale system for large AI workloads |
| GeForce RTX 50 Series | Consumer graphics products based on Blackwell | Desktop and laptop PCs rather than data-center racks |
The hierarchy matters when comparing specifications or prices: a B200 is a GPU, a GB200 is a two-GPU-plus-CPU module, and NVL72 is an entire rack-scale deployment. They are not interchangeable product names.
Rank #2
- Professional GPU with Blackwell Architecture
- Blackwell Architecture
- 24GB GDDR7 with PCIe 5.0 & Ray Tracing
- AI Workstation
What NVIDIA claimed about performance and efficiency
NVIDIA’s launch materials claimed up to 4× faster training and up to 30× faster inference than H100, along with up to 25% lower total cost of ownership and energy consumption for the generative-AI workloads described by NVIDIA.
Those figures are vendor claims, not independent measurements. “Up to” results depend on the workload, software stack, system configuration and comparison baseline. They should not be read as a guaranteed speedup for every model or as a substitute for a benchmark using your own code.
Blackwell for consumers in Q1 2025
At CES on January 6, NVIDIA introduced the GeForce RTX 50 Series, its consumer Blackwell line. The announcement named the RTX 5090, RTX 5080, RTX 5070 Ti and RTX 5070.
NVIDIA specified February availability for the RTX 5070 Ti and RTX 5070, and March–April timing for specified laptop models. The announcement itself does not establish a complete quarter-long record of retail stock, pricing or seller status for every model.
If the question is which physical product to identify, use the exact name NVIDIA GeForce RTX 5090. Retail price and availability are time-sensitive and should be checked at the point of purchase rather than inferred from the launch announcement.
Rank #3
- Form Factor: Plug-in Card
- Cooler Type: Active Cooler
- Maximum Power Consumption: 70W
- Length: 6.6
- Height: 2.7
Choosing an access route
| Decision | Best fit | Main trade-off |
|---|---|---|
| Consumer RTX 5090 versus data-center B200/GB200 | RTX 5090 for a PC or workstation; B200/GB200 for production data-center AI | Consumer cards and data-center systems serve different software, memory, networking, service and deployment needs |
| Own hardware versus cloud rental | Ownership for sustained utilization and control; cloud for rapid access without installing a rack | Ownership brings procurement, power, cooling and operations responsibilities; rental brings recurring usage cost and provider dependence |
| Single-GPU workstation versus NVL72 | Workstation for development, testing or smaller jobs; NVL72 for large distributed training and inference | NVL72 offers rack-scale capacity but requires specialized liquid cooling and facility infrastructure |
| Vendor claims versus independent benchmarks | Use NVIDIA’s figures for the company’s stated workloads; use controlled tests for purchasing decisions | Headline “up to” numbers may not match your model, batch size, precision or software |
When cloud is the sensible first step
GB200 NVL72 cloud instances let a team test Blackwell software and capacity without buying and installing a liquid-cooled rack. Confirm the provider’s instance configuration, scheduling, storage, networking, minimum commitment and billing terms before comparing it with owned hardware.
When ownership may be justified
Owned systems can make sense when utilization is high, data cannot leave your facility, or predictable capacity matters more than avoiding infrastructure work. The relevant cost is total cost of ownership, including electricity, cooling, networking, facilities, support and staffing—not just the accelerator purchase price.
Power, cooling and scale
NVL72 is explicitly a liquid-cooled rack-scale design. That distinguishes it from a conventional desktop or air-cooled workstation: deployment requires compatible facility plumbing, rack integration and operational expertise. The cited NVIDIA materials do not provide a single universal power or cooling figure that applies to every Blackwell configuration, so a buyer should use the exact system vendor’s facility specifications.
What Q1 2025 established—and what it did not
| Question | What can be stated |
|---|---|
| Was Blackwell shipping? | Yes. NVIDIA said production-system shipments began in fiscal Q4 FY2025, the quarter ending January 26, 2025. |
| Was it producing meaningful revenue? | Yes. NVIDIA reported $11.0 billion of Blackwell architecture revenue for that fiscal quarter. |
| Could organizations rent Blackwell capacity? | Yes. NVIDIA announced generally available GB200 NVL72 instances from CoreWeave on February 4, 2025. |
| Was Blackwell available to consumers? | Yes. NVIDIA announced the RTX 50 Series, including RTX 5090, RTX 5080, RTX 5070 Ti and RTX 5070. |
| Is there a complete calendar-Q1 unit-shipment table? | Not stated in the cited NVIDIA materials. |
| Is there a quarter-long retail price and inventory history? | Not established by the launch announcement; prices and stock change by market and seller. |
| Are every performance claim independently validated? | No independent validation is established here; NVIDIA’s headline results remain vendor claims for specified workloads. |
Bottom line for readers evaluating Blackwell
In calendar Q1 2025, Blackwell had crossed from announcement to commercial deployment: production systems were shipping, NVIDIA reported $11.0 billion in Blackwell revenue for its fiscal quarter ending January 26, and CoreWeave offered generally available GB200 NVL72 capacity. The right comparison depends on whether you need a consumer RTX 5090, a B200/GB200 data-center platform, or rented rack-scale capacity. Treat NVIDIA’s speed, inference and efficiency figures as workload-specific vendor claims, and validate costs, cooling and performance against the exact system you plan to run.
Quick Recap
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.
Do these 3 things before closing this tab:
1Clear out junk files and repair common Windows errors2Fix the driver behind crashes, sound loss and screen glitches3Repair Windows errors before they cause bigger problems
