Google TPU IronwoodGoogle
11
Inference-first TPU pods
- Version
- 7th-generation (Ironwood)
- Cost
- from $5.4/unit
- Model
- Gemini
NVIDIA Blackwell
NVIDIA · #5 most active of 18 in Chips & Compute
NVL72 rack-scale coherent GPU domains, the default substrate for frontier training and inference.
100
Trust score · Verified
2 data points checked · 2026-11-08
Current version
GB200 NVL72
Entry cost
Pricing is not published (enterprise/quote-only)
Changes / 30d
5
The short answers · verified November 4, 2026
- What is the latest version of NVIDIA Blackwell?
- The current shipped version of NVIDIA Blackwell is GB200 NVL72, as of November 4, 2026.
- How much does NVIDIA Blackwell cost?
- NVIDIA Blackwell has no published flat entry price. Pricing is not published (enterprise/quote-only)
- What AI model does NVIDIA Blackwell use?
- NVIDIA Blackwell runs primarily on Blackwell.
Source: NVIDIA — www.nvidia.com · Reusable under CC BY 4.0 — cite Tomorrow
Capabilities
- 72-GPU NVLink domain
- FP4 inference
- HBM3E 192GB
- CUDA ecosystem
- Second-generation Transformer Engine
- Fifth-generation NVLink
- FP4 AI Support
- Tensor Cores with microscaling formats
- Liquid-cooled rack-scale design
- NVLink-C2C interconnect
- Confidential Computing
- Multi-Instance GPUs (MIG)
- Tensor Cores
- NVLink
- Transformer Engine
- Multi-Instance GPU (MIG)
- Liquid-cooled design
- CUDA Compatibility
- Second-Generation Transformer Engine
- FP4 and FP8 AI Precision
- Fifth-Generation NVLink
- Tensor Cores with Microscaling Formats
- Liquid-Cooled Rack Design
- Liquid Cooling Design
- NVLink Switch System
- Trillion-Parameter LLM Inference
- FP4 AI
- FP8 precision
- GPU Architecture
- Liquid-cooled
- FP4 AI Precision
- Liquid cooling
- GB200 NVL72
- Grace Blackwell Superchip
- FP4 AI support
- Liquid Cooling
- CUDA Tile C++
- CompileIQ
- CUDA Python 1.0
- FP4 AI inference
- NVLink Interconnect
- CUDA support
- liquid-cooled
- liquid-cooled design
- LLM Inference
- LLM Training
- FP4 AI precision
- FP8 Precision
- Liquid-cooled Design
- Real-Time LLM Inference
- Massive-Scale Training
- NVLink C2C
- NVLink-C2C
- FP4 AI precision support
- FP8 training
- Liquid Cooled
- FP8 AI Precision
- Blackwell Architecture
- FP4 AI Inference
- FP8 Precision Training
- FP8 precision support
- AI Inference
- NVLink interconnect
- GPU architecture
- second-generation Transformer Engine
- fifth-generation NVLink
- GPU Acceleration
- Real-Time Inference
- Accelerated Computing
Pricing
Pricing is not published (enterprise/quote-only)
Entry price over time
Not enough pricing history yet — we start charting from the second observation.
Geek mode
Models under the hood
- disclosed
Blackwell
NVIDIA
- disclosed
Blackwell B200
NVIDIA
- disclosed
Grace CPU
NVIDIA
Context window
—
Public API
no
Multi-model routing
no
Stack signals
- CUDA
- C++
- Python
Shared model stack
Other tracked products running on the same foundation models — a quick read on how much of the catalog moves when one of these models changes.
- BlackwellNVIDIA · 1 product
- Blackwell B200NVIDIA · 1 product
- Grace CPUNVIDIA · 1 product
green disclosed · amber inferred · grey unattributed. Aliases are folded into one model; provider concentration counts only models with an established vendor.
Reported scale
Reference data. Each figure is whatever the source actually said — weekly users, downloads, revenue run-rate — with its own definition and date. These are not comparable between tools and are never used to rank anything.
No public usage figure on record for this product.
Public market scorecard
NVIDIA investor relationsNVIDIA trades as NASDAQ: NVDA. Figures below are as reported for Q1 FY2027 on May 27, 2026; market levels are the close on Aug 6, 2026 and are a reference marker, not a live quote.
Q1 FY2027 revenue
$52B
+42% YoY
- Data center revenue
- ~$46B
- ~88% of total
- Gross margin
- ~74%
- Networking revenue
- ~$6B
- NVLink + Spectrum-X attach
Blackwell/Rubin supply sold out through the year
Also in Chips & Compute
All Chips & ComputeCerebras WSE-3
Cerebras
9
Wafer-scale inference and training
- Version
- CS-4
- Cost
- Free trial credits available; developer tier has free promotional credits.
- Model
- Codex-Spark
Tenstorrent Blackhole
Tenstorrent
8
Open RISC-V AI hardware
- Version
- Blackhole
- Cost
- from $999/unit
- Model
- Blackhole Tensix
7
Cloud-native training silicon
- Version
- Trainium2
- Cost
- Pricing is not published on the page. Available via EC2 instances/UltraServers.
- Model
- Neuron compiler
High-memory GPU alternative
- Version
- MI355X
- Cost
- Enterprise hardware, pricing not published
- Model
- CDNA
Groq LPU
Groq
3
Deterministic low-latency inference
- Version
- GroqCloud API
- Cost
- Pricing not published on current page. Custom/usage rates apply.
- Model
- canopylabs/orpheus-arabic-saudi
Change history
- version
- capability
- version
- version
- capability
New capabilities: GPU Acceleration, Real-Time Inference
66 tracked68 tracked · +GPU Acceleration, Real-Time Inference
source - version
- capability
New capabilities: second-generation Transformer Engine, fifth-generation NVLink
64 tracked66 tracked · +second-generation Transformer Engine, fifth-generation NVLink
source - version
- version
- version
- version
- capability
- version
- version
- version
- version
- version
- version
- capability
- version
- version
- version
- capability
- version
- version
- version
- version
- capability
- capability
New capabilities: FP4 AI Inference, FP8 Precision Training
58 tracked60 tracked · +FP4 AI Inference, FP8 Precision Training
source - version
- capability
- capability
- version
- version
- capability
- version
- capability
- capability
- capability
- version
NVIDIA Blackwell compared
Straight head-to-head pages against the busiest products in Chips & Compute.
How to cite this page
Free to cite and reuse under CC BY 4.0. Permalink: https://tomorrow.aliensquad.ai/tools/nvidia-blackwell
APA
Tomorrow. (2026). NVIDIA Blackwell — version, pricing and model stack [Data set entry]. AlienSquad. Retrieved 2026-11-08, from https://tomorrow.aliensquad.ai/tools/nvidia-blackwell
BibTeX
@misc{tomorrow-tools-nvidia-blackwell,
author = {{Tomorrow}},
title = {NVIDIA Blackwell — version, pricing and model stack},
year = {2026},
publisher = {AlienSquad},
howpublished = {\url{https://tomorrow.aliensquad.ai/tools/nvidia-blackwell}},
note = {Accessed: 2026-11-08}
}