Tesla details Cortex 2 training cluster for FSD and robotics model development

Read the original article →

What happened

Tesla disclosed details on August 19 of its Cortex 2 training cluster in Austin for full self-driving and Optimus robotics models. The facility includes 50,000 in-house AI accelerators optimized for video training workloads.

CEO Elon Musk said Cortex 2 will train unified vision-language-action models shared across automotive and robotics product lines. Production inference remains on dedicated in-vehicle and robot hardware.

Why it matters

Tesla continues vertical integration in AI infrastructure rather than relying solely on cloud providers. Unified models across cars and robots could accelerate data and architecture reuse if the approach works.

Video-heavy training clusters reflect growing demand for multimodal models outside text-centric labs.

MintedBrain take

Cortex 2 is primarily a Tesla product story, not an enterprise AI platform. The relevant takeaway is that video and action-model training is attracting dedicated infrastructure investment industry-wide.

References

This article was originally published at Electrek. For the full piece, read the original article.

Discussion

  • Loading…

← Back to News