OpenAI Says GPT-5.6 Sol Can Autonomously Post-Train
The Decoder · rss · 2026-07-11
The Decoder reports that OpenAI's GPT-5.6 Sol can autonomously post-train a smaller Luna model on a 'fairly vague' prompt.
The report also notes:
- On OpenAI's internal Recursive Self-Improvement (RSI) benchmark, Sol scored 16.2 points higher than GPT-5.5.
- OpenAI believes an 'automated researcher' is becoming achievable.
This is news about model self-improvement capabilities and internal evaluation progress.
More from Models
- Rumor claims Kimi faked performance by serving Claude; DeepSeek new model surprises in evals — realsohamparekh · 2026-09-11
- GPT-5.6 writes well but is instantly forgettable, user complains — BasedRaddka · 2026-09-11
- Opus Refuses Protein Research Codebase Over 'Safety' Concerns, Dev Considers Rolling His Own — josephdviviano · 2026-09-11
- User Hails Unconfirmed 'DeepSeek 4.1 Flash' as an Inflection Point in LLMs — himanshustwts · 2026-09-11
- Terminal Bench v4: GLM-5.3 Leads at 41.9%, Kimi-K3 Underwhelms at 12.6% — Ok_Warning2146 · 2026-09-11
- GPT-6 Astra beats Factorio with enemies in 44 in-game hours at ~$4,500 API cost — liminal_bardo · 2026-09-11