01
Measured, then claimed
We publish numbers we measured, labelled with the hardware. Everything else is a target.
Company
Turbokern exists so that any AI model can run at its best on every NVIDIA GPU without a team of scarce performance experts.
Why we started
Inference is becoming the largest recurring cost of AI. The work that cuts it, profiling, kernels, quantization and serving, is done by a small number of specialists, one model and one GPU at a time.
Each new GPU generation resets that work. We think the loop those engineers run can be automated, measured and repeated for every model, and that the data from each run makes the next one better.
Principles
01
We publish numbers we measured, labelled with the hardware. Everything else is a target.
02
A faster wrong answer is a bug.
03
Outputs are standard NVIDIA artifacts you can inspect and own.
04
A small team of performance, compiler and ML engineers.
Roadmap
Draft roadmap. Dates are targets.
Profiling and the real-GPU benchmark harness for LLM inference on H100.
Kernl, Quantra and Servex with Simyx pruning, run with design partners.
Re-optimization on Blackwell; Edgyn on Jetson Orin.
General availability, on-prem, and training optimization research.
Careers
Draft openings. Email careers@turbokern.com with a note about the hardest performance problem you have solved.
Our stack
Next step
If you have tuned kernels or serving by hand and want to automate it, write to us.