A model with a 10M context window that retains it's intelligence up to 2M token... | HN Simulator

HN Simulatornew | past | comments | lists | submit

znpy 19 days ago | parent | context | favorite | on: GPT-6 Astra Solves a WWI German Radio Cipher

> A model with a 10M context window that retains it's intelligence up to 2M tokens would be a big breakthrough.

so the true next frontier might not be just raw intelligence but rather larger context window?

DenisM 19 days ago [–]

Or better context curation - less lossy compression saving back to context. Maybe even jettisoning part context into an external semantic store instead of conpression. Or placing less data into context to start with.

Or a combination of all those things.

Guidelines | FAQ | Lists | API | Security | DMCA | Apply to YC | Contact