Product
Two hidden API settings tripled GPT-5.6's score on ARC-AGI-3
OpenAI Blog · Jul 29, 2026 · 2 min read
When GPT‑5.6 Sol debuted on the ARC-AGI-3 benchmark, the results were baffling. The same model that solved the cycle do...
Topic: Benchmarking
6 articles
Explore our coverage of Benchmarking — 6 curated articles, summaries, and related resources from the Inblix archive.
Product
Product
AI Meets Life Science
OpenAI Blog · Jul 7, 2026 · 1 min read
LifeSciBench is a new benchmark that tests AI systems' ability to handle complex life science research tasks, going bey...
AI News
AI Benchmarks Fall Short
The Decoder · Jul 3, 2026 · 1 min read
Researchers at the UK's AI Security Institute found that standard benchmarks don't accurately reflect the capabilities...
AI News
AI Model Reaches 56% Solve Rate
The Decoder · Jun 27, 2026 · 1 min read
A new AI benchmark called MirrorCode challenges models to recreate entire programs from scratch without access to the o...
Research
Up to 40% of Arabic AI Benchmarks Are Broken, QIMMA Audit Finds
Hugging Face Blog · Apr 21, 2026 · 2 min read
If you've been tracking Arabic LLM evaluation, you've probably noticed a growing tension: the number of benchmarks and...
Research
VAKRA Benchmark Exposes a Brutal Reality: Top AI Agents Fail at Basic 3-Step Workflows
Hugging Face Blog · Apr 15, 2026 · 2 min read
We've been benchmarking AI agents all wrong. That’s the blunt message from the team behind VAKRA, a new evaluation suit...