18 August 2026

Two AI labs show reasoning and memory boost test performance

• A smaller model from BDH-CQ solved about 30% of difficult reasoning problems at minimal cost per task.
• OpenAI's GPT-5.6 Sol nearly tripled its performance on similar tests by using a memory strategy that reduced output length by six times.
• Both results suggest that how AI systems store and process information matters as much as raw model size.

How it was covered

Latent Spaceswyx & Alessio

BDH-CQ's 150M model using latent-space reasoning achieved 29.5% pass rate on ARC-AGI-1 at $0.0007 per task, while OpenAI reported GPT-5.6 Sol improved from 13.3% to 38.3% on ARC-AGI-3 using retained reasoning and compaction with 6x fewer output tokens, establishing memory strategy as a capability multiplier.