Supercomputers/Systems/CHIE-4
Japan · operational · ai training
CHIE-4
Operated by SoftBank Corp. at SoftBank Corp. (undisclosed site) .
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
- 135.4 PFlop/s
- Rpeak
- 151.9 PFlop/s
- Rmax ÷ Rpeak
- 89.1%
- Cores
- 662,256
- 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 | - NVIDIA DGX SuperPOD (DGX B200) | NVIDIA | - | Reported | top500.org |
| Accelerator | NVIDIA B200 SXM Accelerators CUDA (Blackwell) · TSMC 4NP · 1,000 W NVIDIA B200 SXM 180GB (DGX B200, >4,000 GPUs) | NVIDIA | - | Reported | top500.orgsoftbank.jp |
| CPU | Intel Xeon Platinum 8570 CPUs x86-64 · 56C · 2.1 GHz · Intel 7 Xeon Platinum 8570 56C 2.1GHz | Intel | - | Reported | top500.org |
| Interconnect | NVIDIA InfiniBand NDR (Quantum-2) Interconnect Fat tree or dragonfly+ · 400 Gb/s per port · Under 0.6 microseconds Infiniband NDR400 / NVIDIA Quantum-2 | NVIDIA | - | Reported | top500.org |
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 | 4,000 accelerators | 2025-07 | Operator stated | SoftBank states 'over 4,000' Blackwell GPUs in a DGX B200 SuperPOD (22 Jul 2025). Value is the floor of that statement. | softbank.jp |
The read
SoftBank's Blackwell-generation CHIE cluster (DGX B200), and the one entry in this trio with the clearest independent-benchmark cross-check: SoftBank's own November 2025 release states its HPL-MxP AI-performance rank, a genuinely separate benchmark run at SC25, which happens to agree with TOP500's own concurrently published Rmax rank rather than merely repeating it.
Timeline
- 2025-07 Announced source
- 2025-07 First operational source
- 2026-06 List debut at #21 (June 2026) source
Public list appearances
Pointers only: rank and edition, linking to the canonical entry. We do not reproduce list tables. See sourcing policy.
Site & facility
Change history
- 2026-10-04 Added AI datacenter metrics for chie-4
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 |
|---|---|---|
| top500.org | names system and part | NVIDIA B200 SXM, Intel Xeon Platinum 8570, NVIDIA InfiniBand NDR (Quantum-2) |
| softbank.jp | names system and part | NVIDIA B200 SXM |
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.