Luminal raises $25m Series A to expand inference across different chips
Standard Capital led Luminal’s $25m Series A for its AI inference compiler and cloud.

Choosing a different AI chip can mean changing the software used to run a model. Luminal is building tools to make that choice easier.
What happened
Luminal raised a $25m Series A led by Standard Capital. The company combines a search compiler with an inference cloud. Its compiler explores mathematically equivalent ways to execute a model on a chip and selects an efficient implementation for repeated use.
Why it matters
Hardware options are more useful when developers can deploy their models without rebuilding the surrounding software. A compiler can handle part of the optimisation work between the model and the processor.
The bigger picture
Inference competition involves software as well as chip design. Tools that support different hardware platforms can influence which processors developers are able to use.
