GPT-6 Luna (Max) lands on Agent Arena Pareto frontier at $0.05 median cost per task
arena · x · 2026-09-29
Agent Arena updated its Pareto frontier: OpenAI's GPT-6 Luna (Max) lands with +1.59% net improvement across 8K real-world agentic sessions at a median cost of just $0.05 per task—29% cheaper than DeepSeek-V4.1-Flash and 76% cheaper than Hy4 preview while trailing their scores by 2.27 and 2.51 points. Versus sibling models, it costs 94% less than GPT-6 Sol (Max, $0.82) and 98% less than GPT-6 Astra (Max, $2.59), and sits within 0.09 points of GPT 5.5, a six-place rank jump over GPT-5.6 Luna.
Related event: GPT-6 Luna Hits Agent Arena Pareto Frontier at $0.05 per Task(2 posts)→
More from Models
- Frontier models refuse to harden Windows DCs 43.8% of the time, more if you claim authorization — Aizkmusic · 2026-09-29
- OpenAI DevDay agenda leaks: Codex to get platform capabilities for plugins, agents and apps — testingcatalog · 2026-09-29
- Sonnet 5.5 lands at half Opus 5.5's price, Every team splits after a week of testing — every · 2026-09-29
- OpenAI discloses GPT-6.1 Astra showed deception and overstepped user authorization — soumitrashukla9 · 2026-09-29
- Same Prompt, 21 Replays: Sonnet 5.5 Took the Closest Pokémon 13 Times, Opus 5.5 Never Did — VibeCodyH · 2026-09-29
- ISTA-DASLab's pruned, quantized Qwen3.8 Coder GGUF model trends on Hugging Face — ISTA-DASLab · 2026-09-29