Rebellions

What is Rebellions?

AI inference chips built for efficiency at scale.

Who is Rebellions best for?

Rebellions is best for data-centre and cloud operators that run AI inference at scale and want efficient accelerator hardware.

• Data-centre inference

  RebelServer, RebelRack and RebelPOD scale from one server to a turnkey inference rack and pod on the same software stack.

• LLM serving teams

  The runtime plugs into vLLM, so an existing serving stack runs on Rebellions NPUs without custom integration code.

• ML engineers porting models

  PyTorch, TensorFlow and Hugging Face models compile for the NPU in one step, with no retraining and, for most models, no code changes.

• Kubernetes clusters

  A device plugin and feature discovery let Kubernetes find NPU nodes and schedule inference jobs on them without manual setup.

How much does Rebellions cost?

Rebellions does not publish prices: its inference chips, cards, servers and racks are sold through its sales team and a Let's Talk contact form, and the maker states no free tier.

Rebellions has no pricing page or store: every product, from the Rebel100 chip and RebelCard to RebelServer, RebelRack and RebelPOD, leads to Contact Sales or the Let's Talk form, where you pick a topic such as Business / Product. Plans and current prices are on rebellions.ai.

How do readers rate Rebellions?

No one has rated Rebellions yet. Be the first.

What are the best Rebellions alternatives?

One tool on Recaply does Rebellions's job.

FuriosaAIPaid

An efficient accelerator for LLM inference at 180W.

Questions about Rebellions