[X]score: 0.57

Grok Voice Think Fast 1.0 Hits Pareto Frontier on EVA-Bench

June 10, 2026

xAI's Grok Voice Think Fast 1.0 reaches the Pareto frontier on ServiceNow's EVA-Bench, meaning no evaluated system beats it on both accuracy and experience simultaneously. xAI claims pricing is a fraction of competitors, with the API available at x.ai/api/voice.

HOW THIS AFFECTS YOU

●

builderYou can access Grok Voice via API today with a claimed best-in-class accuracy-experience tradeoff on EVA-Bench, worth benchmarking against ElevenLabs and OpenAI voice APIs on cost.

●

founderCompetitive pricing on a top-performing voice API could shift cost structures for voice agent products — worth evaluating as an alternative to incumbent providers.

read original ↗x.com

← back to feed