shiloz-stack / VeloSpecView on GitHub
Grammar-aware adaptive speculative decoding for structured LLM output — custom CUDA/Triton kernels, 2.7× throughput on enum-heavy schemas
☆17Aug 7, 2026Updated 2 months ago

Alternatives and similar repositories for VeloSpec

Users that are interested in VeloSpec are comparing it to the libraries listed below. We may earn a commission when you buy through links labeled 'Ad' on this page.

Sorting:

Are these results useful?