Speedster7t FPGAs

Speedster®7t FPGAs are optimized for high-bandwidth workloads and eliminate the performance bottlenecks associated with traditional FPGAs.

The Speedster7t FPGA family is optimized for high-bandwidth workloads and eliminates the performance bottlenecks associated with traditional FPGAs. Built on TSMC’s 7nm FinFET process, Speedster7t FPGAs feature a revolutionary new 2D network-on-chip (2D NoC), an array of new machine learning processors (MLPs) optimized for high-bandwidth and artificial intelligence/machine learning (AI/ML) workloads, high-bandwidth GDDR6 interfaces, 400G Ethernet and PCI Express Gen5 ports — all interconnected to deliver ASIC-level performance while retaining the full programmability of FPGAs. Get started today with the VectorPath® accelerator card, featuring the Speedster7t FPGA.

2D Network on Chip (2D NoC) High-Speed Interfaces Machine Learning Processors
  • >20 Tbps bandwidth
  • Interconnects, I/O, memory, internal functional blocks and FPGA fabric.
  • Reduce routing congestion vs. traditional FPGAs
  • Simplify FPGA design
  • PCIe Gen5
  • 112 Gbps SerDes (up to 64)
  • 400G Ethernet
  • GDDR6 memory
    • Up to eight controllers
    • Up to 4 Tbps bandwidth
  • Native support for integer, floating point, bfloat16 and block floating point
  • Speeds up to 750 MHz with co-located block RAM
  • Fully fracturable integer multiplier/accumulator

Target Applications

5G Infrastructure

Computational Storage

Test and Measurement

AI/ML

Networking

Compute Acceleration

VectorPath Accelerator Card

Get started with the VectorPath® accelerator card featuring the Speedster7t FPGA

  • PCIe Gen4 ×16 with qualification path to Gen5
  • 400G and 200G Ethernet connectivity
  • Eight channels of GDDR6 delivering 4 Tbps of memory bandwidth
  • Includes one-year license for Achronix Tool Suite, BittWare BittWorks II toolkit, PCIe drivers

Speedster7t Product Table

Part Number/Name 7t800 7t1500 7t1550 7t3000 7t6000
LUT count (K 6LUTs) 326 692 646 1,300 2,600
Inline Cryptography Yes No Yes Yes Yes
MLP: Multi-fracturable MAC array 864 2,560 880 1,760
LRAM (2.3k) 864 2,560 5,000 10,000
BRAM (73.7k) 1,152 2,560 2,600 5,200
Memory 86 Mb 195 Mb 192 Mb 384 Mb
ML TOps: int8 or block bfloat16 20.5 61 30 61
SerDes 112G/224G 24 32 48 64
DDR4/5 1 DDR5 ×64 (w/ECC) 1 DDR4 ×64 (w/ECC) 2 DDR5 ×64 (w/ECC) 4 DDR5 ×64 (w/ECC)
High Bandwidth Memory Channels 6 GDDR6 (1.5 Tbps,
w/ECC)
16 GDDR6 (4Tbps) 16 GDDR6 @ 4Tbps
(w/ECC)
16 GDDR6 @ 4Tbps
(w/ECC)
PCI Express One ×16
PCie Gen5
One ×8, one ×16
PCie Gen5
Two ×16 with CXL Two ×16 with CXL
Ethernet 8 lanes, 2×400G/
8×100G
16 lanes, 4×400G/
16×100G
16 Lanes
2×800G, 4×400G or
16×100G
32 Lanes
4×800G, 8×400G or
32×100G
2D Network-on-Chip BW – Tbps 12 20 24 48

See Also

Request More Information

Our experts are happy to advise you on how Achronix can help with your toughest design challenges.

Request More Information