- Tesla reboots Dojo 3 after previous supercomputer projects failed to live up to expectations
- AI5 chip performance will reportedly rival Nvidia Hopper while using less power
- Future chips, AI6 and AI7, are planned for incremental technical development
Tesla has restarted development of its Dojo 3 supercomputer project after shelving or abandoning previous versions.
Elon Musk confirmed the move on the X, linking the reboot directly to progress on Tesla’s in-house AI5 chip.
Previous Dojo efforts did not live up to expectations, with Dojo 1 quickly losing relevance Nvidia systems and Dojo 2 will be canceled before completion.
Now that the AI5 chip design is in good shape, Tesla will restart work on the Dojo3. If you’re interested in working on what will be the highest volume chips in the world, send a note to [email protected] with 3 bullet points about the toughest engineering problems you’ve solved.18 January 2026
Tesla reboots Dojo 3 with ambitious internal AI chips
Dojo 3 is framed as a recovery effort rather than a pure breakthrough, as Tesla claims the technical foundation is now strong enough to justify reallocating engineers and capital back to the project.
The Dojo 3 is expected to be Tesla’s first supercomputer built entirely on internal hardware without relying on Nvidia components.
Former Dojo design mixed Tesla silicon with external GPU products that limit differentiation and control, while the new approach brings chip design, system architecture and software under one roof.
Tesla has openly recruited engineers to scale chip production, signaling ambitions for high-volume production.
Central to Dojo 3 is Tesla’s plan to release custom AI chips every nine months, though this is likely to test the company’s resolve.
In terms of application, the AI4 and AI5 chips are linked to self-driving development and humanoid robotics, and the AI6 is linked to Optimus and large data center deployments.
Future iterations, including the AI7, are already mapped out, although expectations point to incremental development rather than radical redesigns.
In addition to vehicles and robots, the supercomputer can support Tesla’s broader ecosystem of AI tools, including training models that compete with established cloud providers.
These claims place Dojo 3 in direct competition with mature AI infrastructure vendors.
According to social commentator Nic Cruz Patane, “Tesla’s chip game is no joke,” noting that the AI5’s performance is roughly comparable to Nvidia’s Hopper on a single chip, approaching Blackwell levels when paired, and running at about 250W compared to the H100’s 700W or Blackwell’s specs at full 000+1.
Tesla claims its chip design delivers similar output at lower watts, but maintaining the planned release cycle will test its discipline and execution consistency.
Its technical promises are ambitious and economically driven, especially given the rising cost of external AI hardware.
Dojo 3 may reduce Tesla’s reliance on third-party silicon, but success will require consistency that previous projects lacked.
Follow TechRadar on Google News and add us as a preferred source to get our expert news, reviews and opinions in your feeds. Be sure to click the Follow button!
And of course you can too follow TechRadar on TikTok for news, reviews, video unboxings, and get regular updates from us on WhatsApp also.
