Moore Threads MTT S5000
Vendor: Moore Threads
Category: GPU Graphics Processor
Architecture: MUSA
Introduction
Moore Threads full-function GPU, using MUSA unified architecture compatible with CUDA ecosystem. Flagship MTT S5000 training-inference unified GPU is in mass production, supporting trillion-parameter large model training. MTT S4000 (FP16 100T, 48GB memory) priced at approximately 30,000 RMB. Listed on STAR Market in December 2025, first quarterly profit in Q1 2026.
Specifications
| Model | Compute | Memory | Interface | TDP | Process |
|---|---|---|---|---|---|
| MTT S5000 | 1,000 TFLOPS (FP8) | 80GB GDDR6 (1.6 TB/s) | OAM | 350W | 12nm |
| MTT S4000 | 128 TFLOPS (BF16) / 256 (INT8) | 48GB GDDR6 (768 GB/s) | PCIe 4.0 | 450W | 12nm |
| MTT S80 | 30 TFLOPS (FP32) | 16GB GDDR6 | PCIe 4.0 | 255W | 12nm |
Official Website
Driver Downloads
Linux
Windows
Related Documentation
OS Support
| Windows | Linux | macOS | Android |
|---|---|---|---|
| ✅ | ✅ | ❌ | ❌ |
Version History
| Version | Release Date | Description |
|---|---|---|
| MUSA 2.0 | 2024 | Domestic graphics rendering + AI compute |
Performance Benchmarks
| Model | Task | Performance Metric |
|---|---|---|
| MTT S5000 | Llama 2 7B Inference | ~40 tok/s (FP16, official data) |
| MTT S4000 | ResNet-50 Inference | ~5000 img/s |
| MTT S80 | Rendering/General Compute | Runs in CUDA-compatible mode |
Pricing
| Model | Reference Price | Notes |
|---|---|---|
| MTT S5000 | Contact vendor | Training-inference flagship |
| MTT S4000 | ~¥30,000 | Inference accelerator |
| MTT S80 | ~¥3,000 | Consumer graphics card |
Quick Installation
Linux (Ubuntu 22.04)
# 1. Install MUSA driver
sudo apt update
sudo dpkg -i mthreads-driver_*.deb
# 2. Install MUSA SDK
tar -xzf musa-sdk-*.tar.gz
cd musa-sdk && sudo ./install.sh
# 3. Verify installation
mthreads-smi
MUSA SDK is downloaded from Moore Threads Developer Center.
Code Examples
Python (MUSA PyTorch)
import torch
# MUDA is Moore Threads' CUDA-compatible backend
assert torch.cuda.is_available(), "Moore Threads GPU not found"
print(f"GPU: {torch.cuda.get_device_name(0)}")
# MUDA API is consistent with CUDA, no code changes needed
x = torch.randn(1024, 1024).cuda()
y = torch.matmul(x, x)
print(f"MUSA matrix multiply: {y.shape}")
Architecture Highlights
- MUSA Unified Architecture: Moore Threads proprietary GPU architecture, compatible with CUDA programming model, supporting CUDA source-level porting
- S5000 Flagship: Designed for large model training and inference, OAM interface suitable for cluster deployment
- Dual-Track Strategy: Simultaneously advancing graphics rendering (MUSA Graphics) and AI computing (MUSA Compute)
Model Compatibility
| Model/Framework | Support | Notes |
|---|---|---|
| PyTorch | ⚠️ MUDA Backend | CUDA-API compatible, requires MUSA PyTorch |
| TensorFlow | ⚠️ Limited | Via MUDA compatibility layer |
| Llama and similar LLMs | ⚠️ | Community adaptation in progress, large model support gradually improving |
| General Inference | ✅ | ResNet/YOLO inference is mature |
| Graphics Rendering | ✅ | DirectX/Vulkan/OpenGL support |
Related Products
If you're evaluating alternatives, the following products may also fit your scenario:
- Jingjia Micro JM9 — Jingjia Micro (GPU Graphics Processor)
- Vastai VA100 — Vastai Semiconductor (GPU Graphics Processor)
- Biren Technology BR100/BR20X — Biren Technology (GPU Graphics Processor)
- MetaX Xiyun C500/C600 — MetaX (GPU Graphics Processor)
- Iluvatar BI-150/BI-V150 — Iluvatar (GPU Graphics Processor)
- Baidu Kunlunxin P800 — Baidu (GPU Graphics Processor)