Meta MTIA 450 vs Meta MTIA 300
The MTIA 450 delivers 5.8x the FP8 throughput of the MTIA 300 (7,000 vs 1,200 TFLOPS dense). The MTIA 450 also carries 72 GB more memory (288 GB vs 216 GB).
Meta MTIA 450
Full specs →Meta MTIA 300
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
The MTIA 450 is the newer part: MTIA, launched in 2027, against the MTIA 300's MTIA from 2026. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.
Dense throughput for the MTIA 450 against the MTIA 300: FP8 7,000 vs 1,200 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
Memory is 288 GB against 216 GB, fed at 18.4 TB/s versus 6.1 TB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.
For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.
Frequently asked questions
Is the MTIA 450 faster than the MTIA 300?
At FP8 precision the MTIA 450 reaches 7,000 TFLOPS dense against 1,200 TFLOPS for the MTIA 300. The performance table on this page lists every published precision for both GPUs.
Which has more memory, the MTIA 450 or the MTIA 300?
The MTIA 450 carries 288 GB of memory versus 216 GB for the MTIA 300. Memory bandwidth is 18.4 TB/s for the MTIA 450 and 6.1 TB/s for the MTIA 300.
How much power do the MTIA 450 and the MTIA 300 draw?
The MTIA 450 is rated at 1.4 kW TDP and the MTIA 300 at 800 W.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.