Flopper.io Flopper.io
EN

Enflame Yunsui i20 vs Enflame Yunsui T20

The Yunsui T20 delivers 5% more FP16 throughput than the Yunsui i20 (134 vs 128 TFLOPS dense). The Yunsui T20 also carries 16 GB more memory (32 GB vs 16 GB).

Yunsui i20: GCU / DTU 2.5, 2021 Yunsui T20: GCU / DTU 2.0, 2021
FP16 dense lead
+5%
Yunsui T20: 134 vs 128 TFLOPS
Memory
16 vs 32 GB
16 GB more for the Yunsui T20
TDP
150 W vs 300 W
Yunsui i20 draws 150 W less
Download comparison sheet

Share this page

Enflame Yunsui i20 vs Enflame Yunsui T20

A post with this link shows the card on the Image tab as its preview.

Post on X Share on LinkedIn
 Printable version →
EF

Enflame Yunsui i20

Full specs →

What is the Enflame Yunsui i20?

The Yunsui i20 is an ASIC from Enflame on the GCU / DTU 2.5 architecture, an accelerator built for AI workloads. It is a PCIe add-in card, so it fits standard servers and workstations. Yunsui i20 full specifications →

Can I rent the Yunsui i20 in the cloud?

Flopper is not tracking any live rental listings for the Yunsui i20 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Yunsui i20 compare with other accelerators?

This page sets the Yunsui i20 against the Yunsui T20. Both parts launched in the same year. Both are Enflame parts, and both ship as a PCIe card, so they fit the same class of system. The compare tool builds the same comparison for any two parts in the catalogue. All GPU comparisons →

Where do the Yunsui i20 figures come from?

Enflame published the Yunsui i20 figures in marketing material rather than a datasheet, and Flopper has no independent confirmation of them. How Flopper verifies specifications →

EF

Enflame Yunsui T20

Leads on FP16 Full specs →

What is the Enflame Yunsui T20?

The Yunsui T20 is an ASIC from Enflame on the GCU / DTU 2.0 architecture, an accelerator built for AI workloads. It is a PCIe add-in card, so it fits standard servers and workstations. Yunsui T20 full specifications →

Can I rent the Yunsui T20 in the cloud?

Flopper is not tracking any live rental listings for the Yunsui T20 yet. The provider directory lists the clouds that rent comparable accelerators on demand. GPU cloud providers →

How does the Yunsui T20 compare with other accelerators?

This page sets the Yunsui T20 against the Yunsui i20. Both parts launched in the same year. Both are Enflame parts, and both ship as a PCIe card, so they fit the same class of system. Flopper also compares it head to head with the L600, across every published precision, memory and power. Yunsui T20 vs L600 →

Where do the Yunsui T20 figures come from?

Enflame published the Yunsui T20 figures in marketing material rather than a datasheet, and Flopper has no independent confirmation of them. How Flopper verifies specifications →

Specifications

EF
Yunsui i20
EF
Yunsui T20
Architecture
GCU / DTU 2.5GCU / DTU 2.0
Launch Year
20212021
Form Factor
PCIePCIe
Memory
16 GB32 GB
Memory Bandwidth
819 GB/s—
TDP
150 W300 W
Process Node
12nmGlobalFoundries 12nm

Performance (TFLOPS)

EF
Yunsui i20
EF
Yunsui T20
No verified data available No verified data available
32 TFLOPS 34 TFLOPS
128 TFLOPS 134 TFLOPS
128 TFLOPS 134 TFLOPS
128 TFLOPS 134 TFLOPS
No verified data available No verified data available
FP6
No verified data available No verified data available
No verified data available No verified data available
256 TOPS 269 TOPS

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

Both launched in 2021, the Yunsui i20 on GCU / DTU 2.5 and the Yunsui T20 on GCU / DTU 2.0.

The Yunsui T20 carries 16 GB more memory (32 GB against 16 GB), so larger models fit on one GPU before you shard.

Power is 150 W against 300 W, and at FP32 the Yunsui i20 does more work per watt. Weight the FP8 and FP16 rows for LLM work and the FP64 row for HPC.

New to the terms? The glossary explains FP32, TF32, BF16, FP16, INT8, Sparsity and TFLOPS.

Frequently asked questions

What is better, the Yunsui i20 or the Yunsui T20?

The Yunsui T20, on throughput. At FP16 precision the Yunsui T20 reaches 134 TFLOPS dense against 128 TFLOPS for the Yunsui i20. The performance table on this page lists every published precision for both GPUs.

Which should you choose, the Yunsui i20 or the Yunsui T20?

Choose the Yunsui T20 for training and high-throughput inference: it has 5% more FP16 throughput than the Yunsui i20. It also carries 16 GB more memory, so larger models fit on one GPU before you shard. The Yunsui T20 draws 150 W more power, so budget for cooling and power delivery before you commit.

Which has more memory, the Yunsui i20 or the Yunsui T20?

The Yunsui T20 carries 32 GB of memory versus 16 GB for the Yunsui i20.

How much power do the Yunsui i20 and the Yunsui T20 draw?

The Yunsui i20 is rated at 150 W TDP and the Yunsui T20 at 300 W. On FP32 throughput per watt, the Yunsui i20 is the more efficient part.

Get Comparison Updates

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

Open in Full Compare Tool → Yunsui i20 full specs Yunsui T20 full specs Estimate cluster cost in the calculator →

© 2026 Flopper.io - Compare the GPUs Powering AI