Flopper.io Flopper.io

NVIDIA GB200 Grace Blackwell Superchip 372GB vs NVIDIA Jetson AGX Thor T5000 128GB

The GB200 delivers 19.3x the FP8 throughput of the Jetson AGX Thor (10,000 vs 517 TFLOPS dense). The GB200 also carries 244 GB more memory (372 GB vs 128 GB).

GB200: Blackwell, 2024 Jetson AGX Thor: Blackwell, 2025
FP8 dense lead
19.3x
GB200: 10,000 vs 517 TFLOPS
Memory
372 vs 128 GB
244 GB more for the GB200
Bandwidth
16.0 TB/s vs 273 GB/s
GB200 moves data faster
Download comparison sheet

Share this page

NVIDIA GB200 Grace Blackwell Superchip 372GB vs NVIDIA Jetson AGX Thor T5000 128GB

A post with this link shows the card on the Image tab as its preview.

Post on X Share on LinkedIn
 Printable version →

NVIDIA GB200 Grace Blackwell Superchip 372GB

Leads on FP8 Full specs →

What is the NVIDIA GB200?

The GB200 is a GPU from NVIDIA on the Blackwell architecture, an accelerator for AI training, inference and high-performance computing. It is a superchip that pairs a CPU and accelerators on one board for rack-scale systems. GB200 full specifications →

Can I rent the GB200 in the cloud?

Flopper is not tracking any live rental listings for the GB200 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the GB200 compare with other accelerators?

This page sets the GB200 against the Jetson AGX Thor. The GB200 is the older generation of the two. Both are NVIDIA parts; the GB200 is a superchip while the Jetson AGX Thor is an embedded module. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the GB200 figures come from?

Every GB200 figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

NVIDIA Jetson AGX Thor T5000 128GB

Full specs →

What is the NVIDIA Jetson AGX Thor?

The Jetson AGX Thor is a GPU from NVIDIA on the Blackwell architecture, an accelerator for AI training, inference and high-performance computing. It is a module for embedded and edge systems rather than a card. Jetson AGX Thor full specifications →

Can I rent the Jetson AGX Thor in the cloud?

Flopper is not tracking any live rental listings for the Jetson AGX Thor yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Jetson AGX Thor compare with other accelerators?

This page sets the Jetson AGX Thor against the GB200. The Jetson AGX Thor is the newer generation of the two. Both are NVIDIA parts; the Jetson AGX Thor is an embedded module while the GB200 is a superchip. Flopper also compares it head to head with the Jetson AGX Thor T4000 64GB, Jetson AGX Orin 64GB and DGX Spark, across every published precision, memory and power. Jetson AGX Thor vs Jetson AGX Thor T4000 64GB →

Where do the Jetson AGX Thor figures come from?

Every Jetson AGX Thor figure on this page traces to NVIDIA's own datasheet or specification table. How Flopper verifies specifications →

Specifications

GB200
Jetson AGX Thor
Architecture
BlackwellBlackwell
Launch Year
20242025
Form Factor
SuperchipModule
Memory
372 GB128 GB
Memory Bandwidth
16.0 TB/s273 GB/s
TDP
—130 W
Process Node
4nm—

Performance (TFLOPS)

GB200
Jetson AGX Thor
80 TFLOPS No verified data available
160 TFLOPS 8.1 TFLOPS
2,500 TFLOPS
5,000 TFLOPS sparse
 No verified data available
5,000 TFLOPS
10,000 TFLOPS sparse
 No verified data available
5,000 TFLOPS
10,000 TFLOPS sparse
 No verified data available
10,000 TFLOPS
20,000 TFLOPS sparse
517 TFLOPS
1,035 TFLOPS sparse
FP6
10,000 TFLOPS
20,000 TFLOPS sparse
 No verified data available
20,000 TFLOPS
40,000 TFLOPS sparse
1,035 TFLOPS
2,070 TFLOPS sparse
10,000 TOPS
20,000 TOPS sparse
517 TOPS
1,035 TOPS sparse

Figures are dense. Where a vendor publishes a figure it sits under the dense one and never decides a winner. Hover any precision for its definition, or read the GPU glossary.

FLOPS by Precision

What actually differs

The Jetson AGX Thor is the newer part: Blackwell (2025) against the GB200's Blackwell (2024).

The GB200 carries 244 GB more memory (372 GB against 128 GB), so larger models fit on one GPU before you shard. Bandwidth is 16.0 TB/s against 273 GB/s, which sets the ceiling on decode speed.

Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP64, FP32, TF32, BF16, FP16, FP8, FP4, INT8, Sparsity and TFLOPS. For how the generations relate, see NVIDIA architectures explained.

Frequently asked questions

Is the GB200 faster than the Jetson AGX Thor?

At FP8 precision the GB200 reaches 10,000 TFLOPS dense against 517 TFLOPS for the Jetson AGX Thor. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the GB200 or the Jetson AGX Thor?

Choose the GB200 for training and high-throughput inference: it has 19.3x the FP8 throughput of the Jetson AGX Thor. It also carries 244 GB more memory, so larger models fit on one GPU before you shard. Packaging differs too: the GB200 is a superchip while the Jetson AGX Thor is an embedded module, so check what your servers accept.

Which has more memory, the GB200 or the Jetson AGX Thor?

The GB200 carries 372 GB of memory versus 128 GB for the Jetson AGX Thor. Memory bandwidth is 16.0 TB/s for the GB200 and 273 GB/s for the Jetson AGX Thor.

Get Comparison Updates

New GPUs added weekly. Be the first to see how they compare.

Open in Full Compare Tool → GB200 full specs Jetson AGX Thor full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI