OpenAI dominates the Output Token per Task Pareto Frontier, says Artificial Analysis
ArtificialAnlys · x · 2026-11-18
Artificial Analysis charts show OpenAI models dominating the Output-Token-per-Task Pareto Frontier, with Claude Fable 5.2 the only non-OpenAI model on it (consuming up to 400k+ output tokens per task at max effort). On the Coding Agent Index (Codex harness, max effort), GPT-6.2 Astra is flat with GPT-6 Astra while GPT-6.2 Sol loses 3 points versus GPT-6.1 Sol.
More from coding & agent
- PentestBench Launches: 17-App Corpus for Testing LLM Web Vulnerability Hunting — hunarbatra · 2026-11-18
- Together AI webinar: how Ricursive trained the local coding model Ripple-65B from scratch — togethercompute · 2026-11-18
- Notre-Dame's south rose window: all 84 panels drawn in code, sandstone layered with shell fragments — Dimillian · 2026-11-18
- Dev recreates Notre-Dame's nave in UE5 with GPT-6.2 Astra, organ track included — Dimillian · 2026-11-18
- GPT-6.2 Sol drops 3 points on Coding Agent Index despite release hype — ArtificialAnlys · 2026-11-18
- Google Maps Platform lands in AI Studio: describe what you want, APIs auto-configured — gaganghotra_ · 2026-11-18