Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading
Paper • 2511.14102 • Published • 18
Cheng Xu
chengxu98
AI & ML interests
None yet
Recent Activity
upvoted a paper about 10 hours ago
Specter: Fast MoE Inference via Quantization-Enabled Multi-Step Speculative Offloading new activity 9 months ago
OpenGVLab/InternVL3_5-GPT-OSS-20B-A4B-Preview:Issue with llama.cpp new activity 9 months ago
OpenGVLab/InternVL3_5-GPT-OSS-20B-A4B-Preview:Issue with llama.cppOrganizations
None yet
upvoted a paper about 10 hours ago
Issue with llama.cpp
18
#3 opened about 1 year ago
by 
ThePun1sher
Issue with llama.cpp
18
#3 opened about 1 year ago
by ThePun1sher