Supercomputers/Systems/Hon Hai Kaohsiung Super Computing Center
Taiwan · under construction · mixed
Hon Hai Kaohsiung Super Computing Center
Also known as Foxconn Kaohsiung GB200 NVL72 supercomputer
Operated by Hon Hai Technology Group (Foxconn) .
Phase 1 seed dataset, compiled by hand. These rows were built from public operator, laboratory and vendor sources. A mechanical second-reader pass has since fetched every cited source: 1092 of 1153 systems have a readable citation that names them, and 40 are genuinely weakly sourced. Every claim carries its source and a confidence tier. Treat anything below verified as a lead, not a citation.
Measured performance
- Rmax
- -
- Rpeak
- -
- Rmax ÷ Rpeak
- -
- Cores
- -
- Power
- -
- Per watt
- - GF/W
Figures are as last publicly reported for the configuration described below, not a live measurement. Where a system was upgraded in place, the post-upgrade configuration is shown and the earlier one appears in the timeline.
What this machine is made of
One row per supplier relationship. “Supplier at build” is the company that shipped the part at the time; where that company has since been acquired, the parent it rolls up to today is shown beside it. That distinction is what makes ticker-level aggregation possible across a thirty-year dataset.
| Role | Part | Supplier at build | Quantity | Confidence | Source |
|---|---|---|---|---|---|
| Integrator | - Built by Ingrasys, Foxconn's data-center equipment subsidiary | Ingrasys 2317TWSE
rolls up to Hon Hai Technology Group (Foxconn) | - | Reported | nextplatform.com |
| Accelerator | NVIDIA GB200 NVL72 Accelerators Arm CPU + CUDA GPU · TSMC 4NP 64 racks of NVIDIA GB200 NVL72, 72 Blackwell GPUs per rack (4,608 GPUs in total) | NVIDIA | 64 reported | Verified | foxconn.comblogs.nvidia.com |
| CPU | NVIDIA Grace CPUs Armv9 (Neoverse V2) · 72C · 3.1 GHz · TSMC 4N 36 NVIDIA Grace CPUs per GB200 NVL72 rack x 64 racks (derived total) | NVIDIA | 2,304 estimated | Reported | blogs.nvidia.com |
| Interconnect | NVIDIA NVLink 5 / NVSwitch Interconnect All-to-all within an NVLink domain · 1.8 TB/s per GPU NVLink Switch joins each 72-GPU rack into one NVLink domain, 130 TB/s per rack | NVIDIA | - | Reported | blogs.nvidia.com |
AI datacenter metrics
Figures beyond an accelerator count, each with who stated it. What can and cannot be compared across AI datacenters.
| Metric | Value | As of | Basis | Note | Source |
|---|---|---|---|---|---|
| Accelerators | 4,608 accelerators | 2024-06 | Design target | 64 GB200 NVL72 racks, completion target 2026 per Hon Hai release (Jun 2024). Plan; no later operator statement fetched. | foxconn.com |
The read
Foxconn's AI computing center in Kaohsiung, announced with NVIDIA on 4 June 2024 and detailed at Hon Hai Tech Day on 7 October 2024: 64 racks of NVIDIA GB200 NVL72 (72 Blackwell GPUs and 36 Grace CPUs per rack, 4,608 GPUs in total), with over 90 exaflops of AI performance claimed (low-precision AI figure, so no FLOPS field is filled). Foxconn and NVIDIA both state the 64-rack, 4,608-GPU configuration. The build is by Foxconn subsidiary Ingrasys per The Next Platform; the first phase was targeted for mid-2025 and full deployment for 2026, for cancer research, LLM development, smart-manufacturing, smart-EV and smart-city work. No later source confirming the current state was found, so status stays under_construction as of the last source date. Network fabric, storage and power are not disclosed.
Timeline
- 2024-06-04 Announced source
Change history
- 2026-10-04 Added AI datacenter metrics for hon-hai-kaohsiung-super-computing-center
- 2026-10-01 Added Hon Hai Kaohsiung Super Computing Center (Foxconn)
Source check
We fetched this system's own citations and recorded whether each page actually mentions it. This is a corroboration signal, not a fact check, and it is published so you can see how well the sourcing holds up rather than take it on trust.
| Cited source | Result | Found on the page |
|---|---|---|
| blogs.nvidia.com | names system, part and count | NVIDIA GB200 NVL72, NVIDIA Grace, NVIDIA NVLink 5 / NVSwitch · counts: 64 |
| foxconn.com | names system, part and count | NVIDIA GB200 NVL72 · counts: 64 |
Checked 2026-10-09 by pnpm verify. Re-run it and the table changes with the web.
Further reading
Inside a node
Why accelerators exist, what the memory hierarchy costs, and how the CPU and accelerator merged onto one package.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.