1.
2.
14× faster embeddings: how we rebuilt the ONNX path in Manticore
(news.ycombinator.com)
3.
Is One Layer Enough? A Single Transformer Layer Matches Full-Parameter RL Train
(news.ycombinator.com)
6.
Generating Images with a 2025 Android
(news.ycombinator.com)
7.
8.
9.
10.
DiffusionBench: Towards Holistic Evaluation of Generative Diffusion Transformers
(news.ycombinator.com)
11.
Krea 2: SOTA open-weights 12B image model
(news.ycombinator.com)
12.
Krea 2 Technical Report
(news.ycombinator.com)
13.
Overfitted a 900KB Transformer to Compress a 100MB CSV into 7MB
(news.ycombinator.com)
14.
Unlimited OCR: One-shot long-horizon parsing
(news.ycombinator.com)
15.
Keanu Reeves Is Helping to Build a New Lego Movie
(gizmodo.com)
16.
LLMs Are Complicated Now
(news.ycombinator.com)
17.
IEEE Rolls Out Large Language Models Virtual Training Course
(spectrum.ieee.org)
18.
Munich 1991: The Roots of the Current AI Boom
(news.ycombinator.com)
19.
Integer Quantization: Deep Dive
(news.ycombinator.com)
20.
GateGPT: 56k tokens per second Transformer (KV cache) on FPGA at 80 MHz
(news.ycombinator.com)
21.
Deficient executive control in transformer attention
(news.ycombinator.com)
22.
Transformers are inherently succinct
(news.ycombinator.com)
23.
24.
1-Bit Bonsai Image 4B Image Generation for Local Devices
(news.ycombinator.com)
25.
CODA: Rewriting Transformer Blocks as GEMM-Epilogue Programs
(news.ycombinator.com)
26.
Show HN: KVBoost – chunk-level KV cache reuse for HuggingFace, 5–48x faster TTFT
(news.ycombinator.com)
27.
28.
When Fast Fourier Transform Meets Transformer for Image Restoration (2024)
(news.ycombinator.com)
29.
When Fast Fourier Transform Meets Transformer for Image Restoration
(news.ycombinator.com)
30.
GenCAD
(news.ycombinator.com)