Claim: OpenAI holds models far stronger than Astra, hinting at early RSI loop
haider1 · x · 2026-09-14
haider1 claims OpenAI already has models much stronger than Google's Astra, and that models smarter than Astra are already helping push through AI research bottlenecks at OpenAI and Anthropic. Citing benchmarks, he says both labs have a clear lead with models above 1T parameters, and suggests recursive self-improvement (RSI) may be one reason. Unverified speculation.
More from Models
- Burkov mocks Meta's Muse model: heavily funded, never tried — burkov · 2026-09-14
- DeepSeek skips V4.1 base model release, drawing criticism over wasted goodwill — teortaxesTex · 2026-09-14
- llama.cpp Adds Maple 20B-A1B Ternary MoE Architecture for CPU and Low-VRAM Devices — jacek2023 · 2026-09-14
- Claude usage boosts quietly removed, fueling talk that 'The Great Compute Crunch has begun' — jacob_posel · 2026-09-14
- PINNACLE benchmark measures real multi-step agentic job completion, tracks fabrication and per-task cost — ryanshrout · 2026-09-14
- Mystery 'antigravity-preview-09-2026' model quietly appears in Google's API — gaganghotra_ · 2026-09-14