GPT-6 Sol and Luna Halve Prices, Sol Cuts Hallucination Rate from 92% to 60%
ArtificialAnlys · x · 2026-09-23
Artificial Analysis publishes a deep dive on GPT-6 Sol and Luna:
- Pricing halved: Sol drops from $4/$20 to $2/$10 per million tokens; Luna from $0.20/$1.20 to $0.10/$0.50. Cost per task falls 50-60% (Sol $1.06 vs $1.99), pushing the cost-efficiency Pareto frontier.
- Coding Agent Index: Sol scores 57 (+2), with Terminal-Bench 4.0 up from 37% to 43%; Luna regresses 2 points to 41.
- Much less hallucination: Sol's rate falls from 92% to 60% (mainly by answering fewer questions — attempt rate 83% vs 99%, accuracy down 5 points); Luna's from 93% to 77%.
- Mixed regressions: GDPval-AA v2.1 drops 100 Elo for Sol and 75 for Luna; manual inspection attributes regressions to shorter deliverables that omit rubric elements.
More from Models
- Claude Code v2.1.280+ lets you switch Opus 5.5 effort mid-session without breaking prompt cache — kimmonismus · 2026-09-23
- Everyone posts 3D render and SVG evals, but nobody shows the refusals and quotas — BLUECOW009 · 2026-09-23
- GPT-6 Astra vs Opus 5.5: Opus Wins Knowledge Work, OpenAI Keeps Hard STEM — johnseach · 2026-09-23
- Polymarket Odds: 91% Chance Claude 6 Ships by End of 2027 — Polymarket · 2026-09-23
- Five Hours of Heavy Opus 5.5 Use on /medium Burned Only 4% of the Weekly Limit — rudrank · 2026-09-23
- One Prompt, Full 3D Scene: GPT-6 Sol Needs Little Inference in 4-Model Test — rohanpaul_ai · 2026-09-23