steven0226/speculative-decoding-bench-rtx4090 download history
steven0226/speculative-decoding-bench-rtx4090 is a dataset on the Hugging Face Hub. In the last 30 days it was downloaded 142 times (46 in the last 7 days), and 646 times in total. It ranks #115,651 among datasets by monthly downloads.
Speculative Decoding Benchmark — RTX 4090 TL;DR: 4,576 benchmark runs measuring speculative decoding speedup / acceptance rate across llama.cpp and LM Studio, Qwen3 (8B/14B) and Llama-3.1-8B target models, on a single consumer RTX 4090 (24GB). Best observed case: the draft-free ngram-
Open steven0226/speculative-decoding-bench-rtx4090 on Hugging Face
Sister project: Paper Pulse, the upvote history of every Hugging Face Daily Paper.