Gemini 4 Hydrogen leads Agent Arena top 10 in task speed
According to Arena, Gemini 4 Hydrogen is the fastest model in its Agent Arena top 10. A typical Hydrogen step takes 7.3 seconds, about 2.7 times faster than Claude Sonnet 5.5 and 3.1 times faster than GPT-6.1 Sol. Arena reports that 82% of Hydrogen sessions finish within 10 minutes, compared with 55% for Sol and 44% for Sonnet 5.5; within 30 minutes, the shares rise to 97%, 81%, and 78%, respectively. Arena also says Hydrogen's net improvement is on par with both models.
Arena’s analysis identifies step latency as a factor in agent-task performance: Hydrogen records shorter steps and higher shares of sessions completed within 10 and 30 minutes while maintaining comparable net improvement.
Gemini 4 Hydrogen by @GoogleDeepMind is the fastest model in the Agent Arena top 10.
A typical Hydrogen step takes 7.3 seconds, about 2.7x faster than Claude Sonnet 5.5 and 3.1x faster than GPT-6.1 Sol.
Faster steps mean less time waiting on tasks. Across Agent Arena sessions, Hydrogen spends under 10 minutes on 82% of its tasks, compared with 55% for GPT-6.1 Sol and 44% for Claude Sonnet 5.5. Within 30 minutes, that rises to 97%, versus 81% for Sol and 78% for Sonnet 5.5.
And it isn't a quality trade-off: Hydrogen's net improvement is on par with both models.
How we measure it: A step is one model API call in an Agent Mode task. All times run from the first model request to the last streamed reply, without tool execution.
#Google DeepMind#Gemini 4 Hydrogen#Claude Sonnet 5.5#GPT-6.1 Sol#Agent Arena