Flopper.io Flopper.io
Meta logo vs

Meta MTIA 400 vs NVIDIA GB200 NVL72 GPU 186GB

The MTIA 400 delivers 20% more FP8 throughput than the GB200 (6,000 vs 5,000 TFLOPS dense). The MTIA 400 also carries 102 GB more memory (288 GB vs 186 GB).

MTIA 400: MTIA, 2026 GB200: Blackwell, 2024
FP8 dense lead
+20%
MTIA 400: 6,000 vs 5,000 TFLOPS
Memory
288 vs 186 GB
102 GB more for the MTIA 400
Bandwidth
9.2 TB/s vs 8.0 TB/s
MTIA 400 moves data faster
See rental pricing from $8.14/hr Download comparison sheet

Share this page

Meta MTIA 400 vs NVIDIA GB200 NVL72 GPU 186GB

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

Post on X Share on LinkedIn
 Printable version →
Meta logo

Meta MTIA 400

Leads on FP8 Full specs →

What is the Meta MTIA 400?

The MTIA 400 is an ASIC from Meta on the MTIA architecture, an accelerator built for AI workloads. It has been announced but is not yet shipping. MTIA 400 full specifications →

Can I rent the MTIA 400 in the cloud?

Flopper is not tracking any live rental listings for the MTIA 400 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the MTIA 400 compare with other accelerators?

This page sets the MTIA 400 against the GB200. The MTIA 400 is the newer generation of the two. This is a cross-vendor matchup, Meta against NVIDIA. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the MTIA 400 figures come from?

Every MTIA 400 figure on this page traces to Meta's own datasheet or specification table. How Flopper verifies specifications →

NVIDIA GB200 NVL72 GPU 186GB

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 an SXM module, the mezzanine format used in dense multi-GPU servers. GB200 full specifications →

Can I rent the GB200 in the cloud?

Yes. Cloud providers tracked by Flopper rent the GB200 by the hour, and the pricing page compares their current rates and the price history side by side. GB200 rental prices →

How does the GB200 compare with other accelerators?

This page sets the GB200 against the MTIA 400. The GB200 is the older generation of the two. This is a cross-vendor matchup, NVIDIA against Meta. Flopper also compares it head to head with the GB300 NVL72 GPU 288GB and Jalapeño, across every published precision, memory and power. GB200 vs GB300 NVL72 GPU 288GB →

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 →

Specifications

Meta logo
MTIA 400
GB200
Architecture
MTIABlackwell
Launch Year
20262024
Form Factor
—SXM
Memory
288 GB186 GB
Memory Bandwidth
9.2 TB/s8.0 TB/s
TDP
1.2 kW—
Process Node
—4nm

Performance (TFLOPS)

Meta logo
MTIA 400
GB200
No verified data available40 TFLOPS
No verified data available80 TFLOPS
No verified data available1,250 TFLOPS
2,500 TFLOPS sparse
3,000 TFLOPS 2,500 TFLOPS
5,000 TFLOPS sparse
No verified data available2,500 TFLOPS
5,000 TFLOPS sparse
6,000 TFLOPS 5,000 TFLOPS
10,000 TFLOPS sparse
FP6
No verified data available5,000 TFLOPS
10,000 TFLOPS sparse
12,000 TFLOPS 10,000 TFLOPS
20,000 TFLOPS sparse
No verified data available5,000 TOPS
10,000 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 MTIA 400 is the newer part: MTIA (2026) against the GB200's Blackwell (2024).

The MTIA 400 carries 102 GB more memory (288 GB against 186 GB), so larger models fit on one GPU before you shard. Bandwidth is 9.2 TB/s against 8.0 TB/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 MTIA 400 faster than the GB200?

At FP8 precision the MTIA 400 reaches 6,000 TFLOPS dense against 5,000 TFLOPS for the GB200. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the MTIA 400 or the GB200?

Choose the MTIA 400 for training and high-throughput inference: it has 20% more FP8 throughput than the GB200. It also carries 102 GB more memory, so larger models fit on one GPU before you shard.

Which has more memory, the MTIA 400 or the GB200?

The MTIA 400 carries 288 GB of memory versus 186 GB for the GB200. Memory bandwidth is 9.2 TB/s for the MTIA 400 and 8.0 TB/s for the GB200.

How much do the MTIA 400 and the GB200 cost to rent?

Live cloud listings tracked by Flopper price no live listings yet for the MTIA 400, and the GB200 from $8.14 per GPU-hour across 3 providers. The rental section on this page links to every provider, rate and the price history for each GPU.

Where to Rent

Meta logo

Meta MTIA 400

No live rental listings tracked for the MTIA 400 yet.

Browse GPU cloud providers

NVIDIA GB200 NVL72 GPU 186GB

Rent from

$8.14 / GPU-hr

Lowest of 3 live listings across 3 providers

See all GB200 rental prices →

Price history and every provider on the pricing page. Some links are affiliate links.

Get Comparison Updates

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

Open in Full Compare Tool → MTIA 400 full specs GB200 full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI