| DeepSeek-V3 Technical Report | Dec 27, 2024 | GPULanguage Modeling | CodeCode Available | 16 | 5 |
| LivePortrait: Efficient Portrait Animation with Stitching and Retargeting Control | Jul 3, 2024 | Computational EfficiencyFace Reenactment | CodeCode Available | 11 | 5 |
| WebLLM: A High-Performance In-Browser LLM Inference Engine | Dec 20, 2024 | CPUGPU | CodeCode Available | 11 | 5 |
| SmolVLA: A Vision-Language-Action Model for Affordable and Efficient Robotics | Jun 2, 2025 | Action GenerationGPU | CodeCode Available | 11 | 5 |
| FlashAttention-3: Fast and Accurate Attention with Asynchrony and Low-precision | Jul 11, 2024 | GPUQuantization | CodeCode Available | 11 | 5 |
| LISA: Layerwise Importance Sampling for Memory-Efficient Large Language Model Fine-Tuning | Mar 26, 2024 | GPUGSM8K | CodeCode Available | 9 | 5 |
| Liger Kernel: Efficient Triton Kernels for LLM Training | Oct 14, 2024 | ChunkingGPU | CodeCode Available | 9 | 5 |
| Divide and Conquer: High-Resolution Industrial Anomaly Detection via Memory Efficient Tiled Ensemble | Mar 7, 2024 | Anomaly DetectionGPU | CodeCode Available | 9 | 5 |
| FlashInfer: Efficient and Customizable Attention Engine for LLM Inference Serving | Jan 2, 2025 | GPUScheduling | CodeCode Available | 9 | 5 |
| Depth Pro: Sharp Monocular Metric Depth in Less Than a Second | Oct 2, 2024 | Depth EstimationGPU | CodeCode Available | 9 | 5 |
