Enflame Blaze T20/T21
Vendor: Enflame Technology
Category: ASIC Dedicated Accelerator
Architecture: Enflame DTU
Introduction
Enflame Technology focuses on cloud AI chips with four generations of proprietary architectures and five chips. Blaze T20/T21 are training and inference products covering chips, accelerator cards, intelligent computing clusters, and a complete software platform. In January 2026, the company received STAR Market IPO acceptance.
Specifications
| Model | Compute | Memory | Interface | TDP | Process |
|---|---|---|---|---|---|
| T21 | 128 TFLOPS (FP16) / 256 (INT8) | 32GB HBM2E (1.6 TB/s) | OAM | 300W | 7nm |
| T20 | 134.4 TFLOPS (FP16) / 268.8 (INT8) | 32GB HBM2E (1.6 TB/s) | PCIe 4.0 | 300W | 7nm |
Official Website
Driver Downloads
Linux
Related Documentation
OS Support
| Windows | Linux | macOS | Android |
|---|---|---|---|
| ❌ | ✅ | ❌ | ❌ |
Version History
| Version | Release Date | Description |
|---|---|---|
| SDK 2.0 | 2024 | Training + inference dual scenarios |
Performance Benchmarks
| Model | Task | Performance Metric |
|---|---|---|
| T21 | Training/Inference | Near A100 level (official data) |
| T20 | Inference | INT8 optimized |
Pricing
| Model | Reference Price | Notes |
|---|---|---|
| T21 | Contact vendor | Enterprise customers |
| T20 | Contact vendor | In mass production |
Quick Installation
Linux
# Install Enflame driver and SDK
sudo rpm -ivh enflame-driver-*.rpm
Code Examples
Python (Enflame)
import torch
# Enflame GPU provides CUDA-compatible backend
assert torch.cuda.is_available()
Architecture Highlights
- DTU Architecture: Enflame proprietary AI accelerator architecture designed for unified training and inference
- Complete Software Stack: Deep learning compiler + runtime + operator library
Model Compatibility
| Model/Framework | Support | Notes |
|---|---|---|
| PyTorch | ⚠️ Compatible | CUDA-compatible backend |
| General AI | ⚠️ | Ecosystem developing |
Related Products
If you're evaluating alternatives, the following products may also fit your scenario:
- Cambricon Siyuan 590 — Cambricon (ASIC Dedicated Accelerator)
- Kunlunxin Gen 2 / Gen 3 — Baidu (GPU Graphics Processor)
- Huawei Ascend — Huawei (NPU Neural Processor)
- Hygon Shensuan DCU — Hygon (ASIC Dedicated Accelerator)
- TsingMicro Reconfigurable RPU — TsingMicro (ASIC Dedicated Accelerator)
- Intel Gaudi (Habana) — Intel (ASIC Dedicated Accelerator)
- NVIDIA GPU / CUDA — NVIDIA (GPU Graphics Processor)