Grok Voice Agent Leads Benchmark Tests While API Costs Surge 60%

Grok Voice Think Fast 2.0 tops agent benchmarks with 56.5% score and 0.7s latency, though API pricing jumps 60% to $0.08/min.

Woofun AI reports that SpaceXAI has released Grok Voice Think Fast 2.0, a voice agent model engineered for customer service, sales, and telephony applications. The system is designed to interpret user input, generate vocal responses, and execute tool invocations, with specific enhancements targeting reasoning, transcription accuracy, and conversational flow.

Independent assessments by Artificial Analysis place the model second in voice synthesis at 82.9%, trailing Qwen Audio 3.0 Realtime Plus.

However, it achieves the highest score in the τ-Voice benchmark for customer service task completion at 56.5%, outperforming competitors. The model initiates audio output within 0.70 seconds.

Concurrently, API pricing has increased by 60%, rising from $0.05 to $0.08 per minute, equating to an hourly cost of $4.80.