Intel cuts Gaudi 3 inference pricing 25% amid competitive GPU market

Read the original article →

What happened

Intel reduced Gaudi 3 cloud inference pricing by 25% on August 12 across partner marketplaces including AWS and Azure. Intel said the move responds to AMD MI450X availability and rising open-model inference demand.

The company published updated performance guides for Llama 4.1 and Nemotron 4 Ultra on Gaudi clusters.

Why it matters

Inference price cuts signal competitive pressure in the AI accelerator market beyond NVIDIA. Lower Gaudi pricing gives cost-sensitive teams another lever when hosting open models.

Updated model guides suggest Intel is optimizing software support for the models enterprises actually deploy.

MintedBrain take

Price cuts help only if performance meets your latency targets. Benchmark Gaudi 3 on your model and batch size before switching from existing GPU contracts.

References

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

Discussion

  • Loading…

← Back to News