Flopper.io Flopper.io
Meta logo

Meta MTIA 300 vs Meta MTIA 200

The MTIA 300 delivers 3.4x the BF16 throughput of the MTIA 200 (600 vs 177 TFLOPS dense). The MTIA 300 also carries 88 GB more memory (216 GB vs 128 GB).

MTIA 300: MTIA, 2026 MTIA 200: MTIA, 2024
BF16 dense lead
3.4x
MTIA 300: 600 vs 177 TFLOPS
Memory
216 vs 128 GB
88 GB more for the MTIA 300
Bandwidth
6.1 TB/s vs 205 GB/s
MTIA 300 moves data faster
TDP
800 W vs 90 W
MTIA 200 draws 710 W less
Download comparison sheet

Share this page

Meta MTIA 300 vs Meta MTIA 200

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 300

Leads on BF16 Full specs →

What is the Meta MTIA 300?

The MTIA 300 is an ASIC from Meta on the MTIA architecture, an accelerator built for AI workloads. MTIA 300 full specifications →

Can I rent the MTIA 300 in the cloud?

Flopper is not tracking any live rental listings for the MTIA 300 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the MTIA 300 compare with other accelerators?

This page sets the MTIA 300 against the MTIA 200. The MTIA 300 is the newer generation of the two. Both are Meta parts. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the MTIA 300 figures come from?

Every MTIA 300 figure on this page traces to Meta's own datasheet or specification table. How Flopper verifies specifications →

Meta logo

Meta MTIA 200

Full specs →

What is the Meta MTIA 200?

The MTIA 200 is an ASIC from Meta on the MTIA architecture, an accelerator built for AI workloads. MTIA 200 full specifications →

Can I rent the MTIA 200 in the cloud?

Flopper is not tracking any live rental listings for the MTIA 200 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the MTIA 200 compare with other accelerators?

This page sets the MTIA 200 against the MTIA 300. The MTIA 200 is the older generation of the two. Both are Meta parts. Flopper also compares it head to head with the MTIA 100, across every published precision, memory and power. MTIA 200 vs MTIA 100 →

Where do the MTIA 200 figures come from?

Every MTIA 200 figure on this page traces to Meta's own datasheet or specification table. How Flopper verifies specifications →

Specifications

Meta logo
MTIA 300
Meta logo
MTIA 200
Architecture
MTIAMTIA
Launch Year
20262024
Form Factor
——
Memory
216 GB128 GB
Memory Bandwidth
6.1 TB/s205 GB/s
TDP
800 W90 W
Process Node
—5nm

Performance (TFLOPS)

Meta logo
MTIA 300
Meta logo
MTIA 200
No verified data available No verified data available
No verified data available No verified data available
No verified data available No verified data available
600 TFLOPS 177 TFLOPS
354 TFLOPS sparse
No verified data available177 TFLOPS
354 TFLOPS sparse
1,200 TFLOPS No verified data available
FP6
No verified data available No verified data available
No verified data available No verified data available
No verified data available354 TOPS
708 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 300 is the newer part: MTIA (2026) against the MTIA 200's MTIA (2024).

The MTIA 300 carries 88 GB more memory (216 GB against 128 GB), so larger models fit on one GPU before you shard. Bandwidth is 6.1 TB/s against 205 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 FP32, BF16, FP16, FP8, INT8, Sparsity and TFLOPS.

Frequently asked questions

Is the MTIA 300 faster than the MTIA 200?

At BF16 precision the MTIA 300 reaches 600 TFLOPS dense against 177 TFLOPS for the MTIA 200. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the MTIA 300 or the MTIA 200?

Choose the MTIA 300 for training and high-throughput inference: it has 3.4x the BF16 throughput of the MTIA 200. It also carries 88 GB more memory, so larger models fit on one GPU before you shard. The MTIA 300 draws 710 W more power, so budget for cooling and power delivery before you commit.

Which has more memory, the MTIA 300 or the MTIA 200?

The MTIA 300 carries 216 GB of memory versus 128 GB for the MTIA 200. Memory bandwidth is 6.1 TB/s for the MTIA 300 and 205 GB/s for the MTIA 200.

How much power do the MTIA 300 and the MTIA 200 draw?

The MTIA 300 is rated at 800 W TDP and the MTIA 200 at 90 W.

Get Comparison Updates

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

Open in Full Compare Tool → MTIA 300 full specs MTIA 200 full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI