GPT-6.2 Astra and Sol land #7 and #9 on Vals Index, Sol costs less than a third per task
rickasaurus · x · 2026-11-18
Vals AI's latest index update: OpenAI's newly released GPT-6.2 Astra and GPT-6.2 Sol rank #7 (67.66%) and #9 (66.06%) — only 1.6 points apart, yet Sol costs less than a third per task ($3.18 vs $10.68).
The top of the chart remains Anthropic/Google territory: Claude Fable 5.2 leads at 71.31% ($17.94/task), followed by Claude Sonnet 5.5, Gemini 4 Carbon, Claude Opus 5.5, and Gemini 4 Argon/Hydrogen. The two GPT-6.2 variants score close overall but diverge more across domains, suggesting different positioning.
More from Models
- GPT-6.2 Astra tops Mercor's OpenWorldEngineer benchmark at 67.2% — morqon · 2026-11-18
- GPT-6.2 Astra takes top spot on OpenAI ECI despite mixed reviews, chart shows — ChrisGPT · 2026-11-18
- Microsoft's MAI models now available via OpenRouter and Vercel, beyond Copilot — marlene_zw · 2026-11-18
- Daniel Miessler walks back GPT-6.2 Astra hype: cheap Fable-parity on benchmarks, but worse in use — daniel_mac8 · 2026-11-18
- OpenAI Questioned for Shipping GPT-6.2 Astra and Sol Despite Tester Backlash — Angaisb_ · 2026-11-18
- Microsoft's MAI-Image-2.7 exits preview, now #2 lab across all three Image Arenas — arena · 2026-11-18