Intel® oneAPI DPC++/C++ Compiler
Talk to fellow users of Intel® oneAPI DPC++/C++ Compiler and companion tools like Intel® oneAPI DPC++ Library, Intel® DPC++ Compatibility Tool, and Intel® Distribution for GDB*
876 Discussions

Request for Intel DLIA (Deep Learning Inference Accelerator) archived software package

heixiu41
New User
10 Views

Hi Intel Support Team,

I am a student currently working on a project involving legacy hardware. I recently acquired an Intel® Deep Learning Inference Accelerator (Intel® DLIA), a PCIe FPGA-based inference card released around 2017, and I am trying to bring it online.

However, the original software package and documentation are no longer available through official channels.

Product details:
- Product Name: Intel® Deep Learning Inference Accelerator
- FPGA: Intel Arria 10
- Reference Document: "Intel® Deep Learning Inference Accelerator Specification and User's Guide" (Revision 1.0, June 2017)
- Original download link (now dead): https://downloadcenter.intel.com/product/122302/

What I need:
1. The installer package: intelDLIA-<version>.tar.bz2
2. FPGA image files (.rpd) and BSP (Board Support Package) drivers
3. Tools: dlia_flash_and_program, dlia_ipselector, dlia_converter
4. The sample applications and pre-trained models mentioned in the user guide

I have already obtained a copy of the user guide from an archive, but without the software stack I cannot proceed with installation and verification.

As a student, I do not have access to enterprise support channels. I would greatly appreciate any of the following:
- A direct download link if the files are still archived internally
- Guidance on where Intel hosts EOL (end-of-life) product software
- Confirmation of whether this product has reached end-of-life with no software retention policy

My goal is purely educational: to understand how early FPGA-based AI acceleration worked and to document the process for other students who may encounter this hardware.

Thank you for your time and assistance.

Best regards,
Zhao Muzhi

0 Kudos
0 Replies
Reply