· ai
Sopro TTS: A 169M model with zero-shot voice cloning that runs on the CPU
Article URL: https://github.com/samuel-vitorino/sopro Comments URL: https://news.ycombinator.com/item?id=46546113 Points: 33 Comments: 10...
-
- · ai
The OptiPFair Series #1: Forging the Future with Small Models — An Architectural Analysis with Pere Martra
Originally published on Principia Agentica The OptiPFair Series – Episode 1 A deep‑dive exploration of Small Language Models SLM optimization. The AI race has...
- · ai
AdaSPEC: Selective Knowledge Distillation for Efficient Speculative Decoders
Introduction AdaSPEC is a new method that speeds up large language models by using a small draft model for the initial generation pass, followed by verificatio...
- · ai
Z-Image GGUF Practical Guide: Unlock Top-Tier AI Art with Consumer GPUs (Beginner Version)
Introduction: Breaking the “GPU Anxiety” – Even 6 GB Can Run Large Models In the world of AI art generation, higher‑quality models usually come with massive si...
- · ai
224× Compression of Llama-70B with Higher Accuracy (Paper and Code)
Article URL: https://zenodo.org/records/17873275 Comments URL: https://news.ycombinator.com/item?id=46212969 Points: 14 Comments: 5...
- · ai
[Paper] CanKD: Cross-Attention-based Non-local operation for Feature-based Knowledge Distillation
We propose Cross-Attention-based Non-local Knowledge Distillation (CanKD), a novel feature-based knowledge distillation framework that leverages cross-attention...