Rising2 sources· last seen 20h ago· first seen 1d ago

OpenAI's GPT-5.6 Sol autonomously post-trained the smaller Luna model with a "fairly underspecified prompt"

According to OpenAI, GPT-5.6 Sol independently fine-tuned the smaller Luna model, triggered by a single "fairly under-specified prompt." In OpenAI's internal RSI benchmark for recursive self-improvement, Sol scores 16.2 points higher than GPT-5.5. OpenAI believes the "automated researcher" is within

Lead: The DecoderBigness: 32openai'sgpt-5solautonomouslypost-trained
📡 Coverage
50
2 news sources
🟠 Hacker News
0
🔴 Reddit
0
📈 Google Trends
0
Full methodology: How scoring works

Receipts (all sources)

According to OpenAI, GPT-5.6 Sol independently fine-tuned the smaller Luna model, triggered by a single "fairly under-specified prompt." In OpenAI's internal RSI benchmark for recursive self-improvement, Sol scores 16.2 points higher than GPT-5.5. OpenAI believes the "automated researcher" is within

The new GPT-5.6 family: Luna, Terra, Sol
RSS · Simon Willison · 1d ago
score 73

Related clusters