Skip to main content
← Lab

holblock

null resultPython · llama.cppInference & servingJuly 8, 2026

the question

Does a long generation head-of-line-block short requests on llama.cpp?

what came out

No -- continuous (iteration-level) batching inflates a concurrent 8-token request by a flat ~1.5x regardless of the long generation's length (128 vs 1024 tokens), versus the 11-52x it would wait if queued behind it. The measured case for iteration-level batching. Pre-registered, independently verified.

method & receipts

  • Result: null result — an honest negative, reported as it came out.
  • Pre-registered the prediction was written down and committed before the run.
  • Reproducible one script re-runs the whole thing from scratch.
  • Independently reviewed an adversarial review pass, kept in the repo.
  • Machine-readable claims every reported number, in a checkable file.
  • Tested — a correctness/benchmark suite ships alongside the code.

→ read the code and re-run it

github.com/v-code01/holblock