Tag

AI benchmarks

Briefings and ideas about AI benchmarks from AI podcast discussions and the research they cite, with links to the original sources.

Latest articles

Models & Research Briefing

Cheap BDH-CQ puzzle solver splits a panel on what follows transformers

BDH-CQ's authors report solving 118 of 400 public ARC-AGI-1 tasks, with up to two answers each, at an estimated $0.00070 inference cost per task. Emad Mostaque welcomed the experiment; Alex doubted it shows progress beyond one benchmark.

Aug 13, 2026 5 min read

Models & Research Briefing

Grok 4.6 nears the leaders as Moonshots panel asks what puts it ahead

xAI's launch table puts Grok 4.6 High at 61 on an intelligence index, level with GPT-5.6 Sol Max. On Moonshots, Alex Wissner-Gross argued, without insider knowledge, that learning from rivals' reasoning traces helps it catch up but not lead.

Aug 13, 2026 6 min read

AI benchmarks

Latest articles

Cheap BDH-CQ puzzle solver splits a panel on what follows transformers

Grok 4.6 nears the leaders as Moonshots panel asks what puts it ahead

Help us understand what gets read