Efficient and Lossless Moe Diffusion LLM Inference with I/O-Aware Expert Offload
1 points, 1 comments on Hacker News
🏷️ #efficient
5 headlinesCodegenBench: Can LLMs Write Efficient Code Across Architectures?
1 points, 0 comments on Hacker News
Efficient Training on Multiple Consumer GPUs with RoundPipe
2 points, 0 comments on Hacker News
AV2 video codec released: 30% more efficient than AV1
2 points, 1 comments on Hacker News
Trace-Based Adaptive Cost-Efficient Routing
2 points, 1 comments on Hacker News