SpaceXAI's Grok Voice Think Fast 2.0 Jumps to #2 on Speech Benchmarks, Cuts Response Time in Half
SpaceXAI released Grok Voice Think Fast 2.0 on July 29, lifting its score on Artificial Analysis's Speech to Speech Index to 82.9% and its time-to-first-audio to 0.70 seconds, ahead of OpenAI's GPT-Realtime-2.1 and Google's Gemini 3.1 Flash on the same tests.
SpaceXAI, the combined xAI-SpaceX entity, released Grok Voice Think Fast 2.0 on July 29, 2026, a successor to the Grok Voice Think Fast 1.0 model it shipped earlier this year, with sizable gains on independent speech-to-speech benchmarks and a sharp cut in latency.
Benchmark gains
According to results published by Artificial Analysis, Grok Voice Think Fast 2.0's "High" reasoning variant scored 82.9% on the firm's Speech to Speech Index, debuting at #2 overall — behind only Qwen Audio 3.0 Realtime Plus (84.1%) and ahead of OpenAI's GPT-Realtime-2.1 High (79.1%) and Google's Gemini 3.1 Flash (69.5%). That's a 7.3 percentage-point jump from Think Fast 1.0's 75.7% score. On Artificial Analysis's Tau Voice benchmark, which measures agentic voice performance, the new model took the top spot at 56.5%, ahead of Qwen Audio 3.0 Realtime Plus (54.6%) and its own predecessor (52.1%).
Faster and among the first to reply
Beyond raw accuracy, SpaceXAI emphasized latency: average time-to-first-audio fell from 1.25 seconds in version 1.0 to 0.70 seconds in version 2.0, making it, per Artificial Analysis, the only model in the Speech to Speech Index's top five averaging under one second to first response. The company also said reasoning-token usage dropped by roughly 60% version-over-version, which should lower the effective cost of running the model at scale even before accounting for its published price of $0.08 per minute of audio.
Why it matters
Voice has become one of the more competitive fronts in the model race this year, as OpenAI, Google and Alibaba's Qwen team all ship realtime speech models aimed at voice agents and assistants rather than text chat. SpaceXAI is positioning Think Fast 2.0 specifically for agentic use inside its Agent Builder tooling, not just as a conversational demo, which tracks with the company's broader push — following July's Grok 4.5 release — to compete on task-completion and latency in agent workflows rather than general chatbot benchmarks alone. SpaceXAI said it will automatically migrate existing Grok users from Think Fast 1.0 to the new model on August 5, 2026.
Sources
- Artificial Analysis benchmark results for Grok Voice Think Fast 2.0
- SpaceXAI launches Grok Voice Think Fast 2.0 — americanbazaaronline.com
- SpaceXAI launches Grok Voice Think Fast 2.0 on Agent Builder — TestingCatalog
- xAI Unveils Voice AI 'Grok Voice Think Fast 2.0' with Dramatically Improved Transcription Accuracy and Inference Speed — BigGo Finance
AI-assisted reporting, overseen by the AgentsAI team. Spotted an error? Let us know.
Related agents
More ai news
Sanders and Casar Introduce Bill to Ban 'Artificial Superintelligence' Outright
Sen. Bernie Sanders and Rep. Greg Casar introduced the Ban Artificial Superintelligence Act on September 3, which would permanently prohibit superintelligent AI systems, pause frontier development pending federal safety rules, and impose prison terms and a 'corporate death penalty' for violations.
Anthropic Says Claude Produced the First Machine-Checked Proof of Fermat's Last Theorem
Working largely autonomously for 11 days on the open Prove2Me platform, Claude generated a 13-million-line Lean formalization of Fermat's Last Theorem, which mathematician Kevin Buzzard called an 'extraordinary autoformalization achievement.'
OpenAI Launches GPT-6 Astra, Its First Model Rated 'Critical' Risk for Cyber Offense
OpenAI says Astra is state-of-the-art on coding, computer use and science benchmarks, and is the first model to trip the company's 'critical' cybersecurity threshold — able to find and exploit unknown vulnerabilities without step-by-step human guidance.
Nvidia Agrees to Acquire Hugging Face for $11.9 Billion, Confirming Weeks of Sale Talk
Nvidia will pay $11.9 billion in cash for the open-source AI hub, plus up to $1 billion in retention equity for employees who join, confirming reports from late August that Hugging Face was exploring a sale.