← The AI Systems Map
L0 · LabsBenchmark / eval

Evaluation & benchmark infrastructure

MMLU/GPQA/SWE-bench/ARC-AGI and eval harnesses

Benchmarks (MMLU, GPQA, SWE-bench, ARC-AGI-2, FrontierMath, HLE) and harnesses (lm-eval-harness, HELM, Chatbot Arena/LMArena) are how model capability is measured, purchased and regulated. Benchmark saturation and contamination make eval design and private held-out sets a strategic, contested layer of the ecosystem.

Benchmarks

SWE-bench, ARC-AGI-2, FrontierMath

Arenas

LMArena/Chatbot Arena

How it fits the stack

Evaluation & benchmark infrastructure with what it depends on (above) and what it feeds (below). The figure renders as a crawlable diagram and upgrades to an interactive 3D graph as it scrolls into view.

suppliesused byused byEvaluation & benchmarkinfrastructureLabsAI Safety Institutes(US CAISI / UK AISI)Claude (Opus / Sonnet)GPT-5 / o-series
Evaluation & benchmark infrastructureFeeds ↓

Evaluation & benchmark infrastructure in the AI stack. Evaluation & benchmark infrastructure with its immediate upstream dependencies (top) and downstream dependents (bottom) in the AI value chain. Hover a node in 3D, or read the full relationships below.

Graph data (text) — 4 entities, 3 relationships
  • AI Safety Institutes (US CAISI / UK AISI)suppliesEvaluation & benchmark infrastructure
  • Claude (Opus / Sonnet)used byEvaluation & benchmark infrastructure
  • GPT-5 / o-seriesused byEvaluation & benchmark infrastructure