Models and infrastructure
Grok combines language models with voice conversations and image and video generation. xAI trains the models on its Colossus computing clusters, which ended 2025 with more than one million H100 GPU equivalents.
What comes next
At the announcement, Grok 5 was in training. The company also offered a voice-agent API for developers, extending its speech technology beyond the Grok app and Tesla vehicles.
Where the funding goes
- Expand computing infrastructure.
- Develop and deploy new AI products and fund model research.