Moore Threads MTT S80 vs Moore Threads MTT S3000 32GB
The MTT S3000 delivers 5% more FP32 throughput than the MTT S80 (16 vs 15 TFLOPS dense). The MTT S3000 also carries 16 GB more memory (32 GB vs 16 GB).
Moore Threads MTT S80
Full specs →Moore Threads MTT S3000 32GB
Full specs →Specifications
Performance (TFLOPS)
FLOPS by Precision
What actually differs
The MTT S80 is the newer part: MUSA ChunXiao, launched in 2024, against the MTT S3000's MUSA ChunXiao from 2023. 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 MTT S80 against the MTT S3000: FP32 15 vs 16 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.
Memory is 16 GB against 32 GB, fed at 448 GB/s versus 448 GB/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 MTT S80 faster than the MTT S3000?
At FP32 precision the MTT S3000 reaches 16 TFLOPS dense against 15 TFLOPS for the MTT S80. The performance table on this page lists every published precision for both GPUs.
Which has more memory, the MTT S80 or the MTT S3000?
The MTT S3000 carries 32 GB of VRAM versus 16 GB for the MTT S80. Memory bandwidth is 448 GB/s for the MTT S80 and 448 GB/s for the MTT S3000.
Get Comparison Updates
New GPUs added weekly. Be the first to see how they compare.