Google's Suspected Gemini 4 Pro AI System Surpasses Top Models in Leaked Benchmark Results
Google's next flagship AI system, suspected to be Gemini 4 Pro, has appeared on the Arena benchmarking platform under the name 'gemini-3.8-flash.'
The model carries the internal codename Argon and has not been officially announced by Google.
Leaked benchmark results suggest it outperformed OpenAI's GPT-6 Astra and Anthropic's Claude Fable 5.1 in coding, reasoning, and agentic tasks.
The model scored 88% on DeepSWE v1.1 for AI-agent coding tasks, 95.3% on Terminal-bench 2.1 for encoding capability, and 86.8% on OSWorld-2.0 for computer proficiency.