Show HN: Maple – Sub-2-BPW Moe Pruning to Run Gemma 4 26B at 45 TPS on a 6GB GPU
A technique for improving the efficiency of large language models (LLMs) on GPUs.
Hacker News submitter: AmSach
Submitted to Hacker News:
A technique for improving the efficiency of large language models (LLMs) on GPUs.
Hacker News submitter: AmSach
Submitted to Hacker News: