Arena.ai launched its new Agent Arena Index, scoring OpenAI's freshly released GPT-6.2 Astra and Sol on 7,100+ real agentic sessions. Astra ranks #6 at 1291 points with a 40% lower median cost per task, while Sol ranks #11 at 1241 points and joins the cost Pareto frontier at 69% lower cost than Astra.
Agent Arena Index
Composite index evaluating agent performance on complex workflows, tool use and feedback adaptation using millions of real task
Curated Intelligence
Google DeepMind has shipped four Gemini 4 models since September 30, compressing release gaps from 28 days to just 6. The latest, Gemini 4 Neon (High), ranks #2 overall on the Agent Arena Index with 1336 points at a $2.82 median cost per task, scoring above Claude Opus 5.5 while costing 53% less and 69% less than Claude Fable 5.2.
Arena.ai has launched the Agent Arena Index with six focused sub-indexes plus an Alignment Index, built on millions of real-world tasks and traces from its global community. The benchmarks measure how well agents complete complex workflows, use tools and adapt to feedback, while the Alignment Index gauges fidelity to user intent.