SpecFit: Not All Draft Tokens Cost the Same in Offloaded MoE Inference
Mixture-of-experts (MoE) models activate only a few experts per token, but when their expert weights are offloaded to host memory, host-to-device (H2D) expert transfers enter the critical path of decoding. Speculative decoding verifies several draft tokens in one forward pass, yet in an MoE model each verified token activates its own experts, so a larger draft tree transfers more experts. Because of this H2D cost, speculative decoding commits several tokens per verification cycle yet is sometimes even slower than autoregressive decoding. Based on this observation, we propose SpecFit, which chooses the verified draft subtree to maximize predicted throughput rather than acceptance length. S...
| Authors | Hongliang Cao, Xiuyu Li, Siqi Wang et al. |
|---|---|
| Primary category | cs.AI |
| Published | 11 October 2026 |
| Last updated at source | 11 October 2026 |
Silytics shows public metadata from arXiv and links to the original. Information may be incomplete or out of date. Last refreshed 14 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.