OpenAI’s September 2026 releases of GPT-6 Sol, Luna, and especially the September 29 GPT-6.1 Sol update have positioned the Sol line as the efficiency-focused tier within the GPT-6 family, building directly on Astra’s capabilities for coding, computer use, and professional tasks at lower cost. Recent HLE leaderboards show GPT-6 Astra debuting near 54.7% on verified variants and GPT-6.1 Sol scoring around 52.9%, reflecting steady gains from earlier GPT-5.6 models but trailing Anthropic’s Claude Opus 5.5 and Fable 5.1, which lead at 61-65% across tool-assisted and closed-book evaluations. Traders are watching whether OpenAI’s next Sol iteration clears higher thresholds on this expert-level benchmark before the June 2027 resolution window, amid rapid iteration cycles and competitive pressure from Google’s Gemini and Meta’s Muse lines.