Isambard-AI phase 2
Isambard-AI is the UK's national AI supercomputer, run by the University of Bristol's Bristol Centre for Supercomputing with HPE, used for large-scale model training, inference and AI research in health, climate and robotics.
What it runs and how to get time
From the operator- Used for What the operator says the system is used for: science domains, missions or workloads. Taken from the operator, its funder or the system vendor, not from TOP500.
- Large-scale AI model training and inference, UK language models, protein modelling, medical imaging, climate and pollution modelling, robotics and drug discovery. 4
- How to get time How researchers or customers get time on the system, naming the allocation programme where the operator states one.
- Allocated externally: UKRI AI Research Resource (AIRR) calls, a University of Bristol rolling call for Bristol staff, and the Sovereign AI route for UK startups. 6
- Cooling Cooling design stated by the operator or supplier.
- Fan-less direct liquid cooling: a closed coolant/water loop through the blades; cuts cooling power by up to 90%. 4
- Power design Facility power, energy supply or heat reuse as stated by the operator. Not the benchmark power in the Power column.
- Runs on zero-carbon UK renewable electricity; PUE about 1.08; waste heat could be recycled for nearby homes and businesses. 4
- Facility Building, data centre or campus that houses the system, as described by the operator.
- Modular data centre of four container-style units at the NCC, Bristol and Bath Science Park, near Bristol. 4
- Upgrade or successor Announced upgrade, expansion, successor or retirement plan, with dates where the operator gives them.
- BriCS is adding an inference service and the BRAID data storage facility, and is planning a successor system; no dates given. 7
- Production start Month the operator says the system entered production.
- 2025-08 4
- Status Operating status, filled only when the operator states it.
- In production 4
Benchmark and hardware
Source per value- Rank Position in the TOP500 edition named in the Edition column. Historic systems keep the rank of the edition where they last appeared.
- 13 1
- Rmax PFLOP/s Measured double-precision LINPACK (HPL) performance of the listed configuration. Older TFLOP/s and GFLOP/s figures are converted to PFLOP/s.
- 216.5 1
- Rpeak PFLOP/s Theoretical peak double-precision performance of the listed configuration. This is not AI or mixed-precision throughput.
- 278.58 1
- HPL efficiency % Rmax divided by Rpeak: the share of theoretical peak the system delivered on LINPACK. Computed from the two sourced figures.
- 77.7 1
- Site Host or operator named by TOP500 for the installation.
- University of Bristol 1
- Country Country or region reported by TOP500 for the site.
- United Kingdom 1
- City Site city as reported by TOP500 or the operator. It can be the operator address rather than the machine room.
- Bristol 3
- Region World region of the installation country (grouping applied by this site).
- Europe 1
- Vendor Manufacturer or system supplier exactly as submitted to TOP500.
- HPE 1
- Vendor group Supplier grouped for filtering. Bull is the Eviden/Atos line; Other covers single-system suppliers.
- HPE 1
- System model Hardware platform in the cited TOP500 edition.
- HPE Cray EX254n 1
- Processor CPU description in the cited TOP500 edition; clock and core count refer to the processor model.
- NVIDIA Grace 72C 3.1GHz 1
- Accelerator Accelerator model listed for the system. Empty when TOP500 lists none.
- NVIDIA GH200 Superchip 1
- Accelerator vendor Maker of the accelerator. None listed means TOP500 names no accelerator for the system.
- NVIDIA 1
- Interconnect Network fabric exactly as listed in the cited TOP500 edition.
- Slingshot-11 1
- Fabric Interconnect family derived from the listed interconnect (grouping applied by this site).
- Slingshot 1
- Cores cores Total cores submitted to TOP500, including accelerator cores where TOP500 counts them. Not a CPU-only count and not a CUDA-core count.
- 1,028,160 1
- Installed Installation year reported in the TOP500 system record. Not necessarily the year of production acceptance.
- 2025 2
- Edition TOP500 edition (YYYY-MM) that supplies the ranking and benchmark data. A historic edition does not assert current operating status.
- 2026-06 1
- OS Operating system from the TOP500 system record.
- HPE Cray OS 2
- TOP500 record Link to the TOP500 database record for this exact submission.
- https://www.top500.org/system/180388/ 2
- Nodes nodes Compute node count as described by the operator. It can exceed the nodes used in the benchmark run.
- 1,362 4
- Accelerators per node per node Accelerator packages or superchips per node, as the operator describes them.
- 4 4
Sources
- TOP500 list, June 2026 accessed 2026-11-11
- TOP500 system record accessed 2026-11-11
- TOP500 site record accessed 2026-11-11
- bristol.ac.uk (operator page) accessed 2026-11-11
- July: UK's most powerful supercomputer launches in Bristol | University of Bristol accessed 2026-11-11
- Applying for Access (Bristol Centre for Supercomputing Documentation) accessed 2026-11-11
- 2026: One year of Isambard-AI | Research | University of Bristol accessed 2026-11-11