GitHub ↗ TMMLU+ ↗
FoxBrain EvalHub September 8, 2026 • Latest: GPT-6 Astra (Sep 3, OpenAI)

Frontier Benchmark & Model Intelligence — HHRI-AI Research
A curated, up-to-date reference of benchmarks and frontier models used by leading AI labs. Use this to select evaluation targets for FoxBrain model assessments across 12 capability domains.

Last updated: September 8, 2026  •  Added OpenAI's GPT-6 Astra (Sep 3), Google's Gemini 3.8 Flash & Flash Cyber, Meta's Muse Spark 1.3 (Sep 2) — a dense wave of 4 frontier launches within 72 hours
70
Benchmarks
9
Domains
77
Models
19
Sources
Domain:
Benchmark Domain Priority Metric Used by labs Description