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

https://the-decoder.com/wp-content/uploads/2026/07/gpt_56_sol.png" style="height: auto; margin-bottom: 10px;" width="1376" />

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 reach.

The article https://the-decoder.com/openais-gpt-5-6-sol-autonomously-post-trained-the-smaller-luna-model-with-a-fairly-underspecified-prompt/">OpenAI's GPT-5.6 Sol autonomously post-trained the smaller Luna model with a "fairly underspecified prompt" appeared first on https://the-decoder.com">The Decoder.