Speculative Decoding Acceptance Rate: My LLM Got 47% Slower
This story is from 2026-09-16. It is preserved in the archive; the latest stories are on the live feed.
I turned on speculative decoding on a Friday afternoon expecting a free 2x. My local 32B model went from 34 tokens/sec to 18. Not noise. Not a warmup artifact. Consistently, reproducibly, half speed. I had added a second model to my GPU, burned VRAM for it, and made everything worse. The thing nobo…
Read the full story at DEV Community — Machine Learning ↗
Timeline · 1 report
- 2026-09-16 04:58 · DEV Community — Machine Learning
Speculative Decoding Acceptance Rate: My LLM Got 47% Slower