UK universities release TEASBench: measuring AI inference cost and performance across hardware
PontiEdoardo · x · 2026-09-01
Developed at Edinburgh, EPCC and Imperial College under ARIA's "Scaling compute" programme, TEASBench is an open benchmark suite measuring cost, accuracy, and performance of AI inference on diverse hardware. Unlike benchmarks with fixed context lengths, it doesn't constrain input/output tokens, exposing real-world hardware limits, and also reports energy and user-experience estimates. It evolves on a 6-monthly cycle — the August 2026 release covers sparse MoE and reasoning models — with NVIDIA supported first and AMD/Tenstorrent pipelines coming.
More from Infra
- Auction-based LLM routing paper finds fixed token pricing wastes 10%-71% in costs — _onionesque · 2026-10-18
- Running Ling 3.1 Flash (560B MoE, 25B Active) Across 4 DGX Sparks via llama.cpp RPC — BerryBlush_Rogers691 · 2026-10-18
- SGLang Summit sponsors jump from 11 to 36 in one week — ying11231 · 2026-10-18
- Oracle CDS hits record 300bps, called first AI trade domino after Barclays sell downgrade — arthurcolle · 2026-10-18
- Agent Coding Workflow Would Cost ~$40k/Month on API Pricing, Dev Reveals — ChrisRackauckas · 2026-10-18
- PS5LM: llama.cpp port turns a jailbroken PS5 into a home LLM server (500+ GitHub stars) — wateriscoding · 2026-10-18