Supercomputers/Systems/Kyber
France · operational · ai training
Kyber
Also known as InstaDeep Kyber
Operated by InstaDeep at Equinix Saint-Denis (Kyber) .
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 | - Dell PowerEdge XE-series servers (Dell AI Factory with NVIDIA) | Dell Technologies | - | Reported | delltechnologies.comconstellationr.com |
| Accelerator | NVIDIA H100 SXM5 Accelerators CUDA (Hopper GH100) · 16,896C · 1.98 GHz · TSMC 4N · 700 W 224 NVIDIA H100 GPUs (count per Constellation Research; InstaDeep confirms H100 and ~0.5 exaFLOPs FP16) | NVIDIA | 224 reported | Reported | constellationr.com |
| Interconnect | - NVIDIA Spectrum networking; 400 Gbps RoCE network | NVIDIA | - | Reported | delltechnologies.comconstellationr.com |
The read
Near-exascale NVIDIA H100 cluster built and managed in-house by InstaDeep (BioNTech's AI subsidiary) in an Equinix data centre in Saint-Denis near Paris, unveiled 24 October 2024 alongside BioNTech's AI Day and used for biology foundation models (protein and DNA language models for BioNTech) as well as industrial AI. InstaDeep states H100 GPUs, about 0.5 exaFLOPs in FP16 and a tenfold increase in its compute, and Dell's case study confirms Dell PowerEdge XE-series servers with NVIDIA Spectrum networking. The specific figures of 224 H100 GPUs, 86,000 CPU cores, 1.7 petabytes of storage and a 400 Gbps RoCE network come from Constellation Research's account of the BioNTech AI Day and are not restated on InstaDeep's own page, so they are reported only. The 0.5 exaFLOP figure is a mixed-precision AI number, not HPL, so no FLOPS field is recorded. Included as a biotech-affiliated system: InstaDeep also serves non-biology customers.
Timeline
- 2024-10 Announced source
Site & facility
Change history
- 2026-10-01 Added Kyber (InstaDeep / BioNTech, France)
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 |
|---|---|---|
| instadeep.com | names system and part | NVIDIA H100 SXM5 |
| constellationr.com | names system, part and count | NVIDIA H100 SXM5 · counts: 224 |
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.
Operating a supercomputer
Acceptance testing, why "installed" and "in production" are six months apart, and the machine lifecycle.
The interconnect
Topologies, why latency and tail behaviour matter more than bandwidth, and how the fabric market consolidated.