| RealisVSR: Detail-enhanced Diffusion for Real-World 4K Video Super-Resolution | | 25.84 | 0.1935 | 0.7601 | | | | | RealisVSR | 2025-07-25 |
| Upscale-A-Video: Temporal-Consistent Diffusion Model for Real-World Video Super-Resolution | ✓ Link | 25.83 | 0.268 | 0.733 | | 0.737 | | | Upscale-A-Video | 2023-12-11 |
| TRaM-VSR: Importance-Aware Token Routing and Merging for One-Step Diffusion Video Super-Resolution | ✓ Link | 24.38 | 0.2903 | 0.6856 | 0.4676 | 1.7 | | | TRaM-VSR | 2026-07-24 |
| DOVE: Efficient One-Step Diffusion Model for Real-World Video Super-Resolution | ✓ Link | 24.30 | 0.2997 | 0.6740 | 0.4985 | 2.05 | | 0.8574 | DOVE | 2025-05-22 |
| DAM-VSR: Disentanglement of Appearance and Motion for Video Super-Resolution | | 24.246 | 0.367 | 0.668 | | 1.769 | | | DAM-VSR (ResShift ISR) (extra data) | 2025-07-01 |
| Towards Redundancy Reduction in Diffusion Models for Efficient Video Super-Resolution | | 23.75 | 0.2608 | 0.6417 | 0.4817 | 2.74 | | 0.8700 | OASIS (extra data) | 2025-09-28 |
| DiffVSR: Enhancing Real-World Video Super-Resolution with Diffusion Models for Advanced Visual Quality and Temporal Consistency | | 23.713 | | | 0.785 | 1.492 | 68.040 | 0.878 | DiffVSR (extra data) | 2025-01-17 |
| FlashVSR+: Efficient Video Super-Resolution via Hybrid Attention and Reward-Yield Distribution Matching | ✓ Link | 23.64 | 0.2698 | 0.625 | | | 64.71 | | FlashVSR+ (Ours-Base) | 2026-10-13 |
| FlashVSR: Towards Real-Time Diffusion-Based Streaming Video Super-Resolution | ✓ Link | 23.31 | 0.3866 | 0.6110 | | | | | FlashVSR (Tiny) | 2025-10-14 |
| DUO-VSR: Dual-Stream Distillation for One-Step Video Super-Resolution | | 22.96 | 0.289 | 0.674 | 0.4222 | 1.98 | 65.24 | 0.8728 | DUO-VSR (extra data) | 2026-03-23 |
| Asymmetric VAE for One-Step Video Super-Resolution Acceleration | | 22.85 | 0.3513 | 0.6614 | 0.5888 | 1.90 | 65.38 | 0.8558 | FastVSR (f8 encoder / f16 decoder) (extra data) | 2025-09-29 |
| SwiftVR: Real-Time One-Step Generative Video Restoration | ✓ Link | 22.67 | 0.2947 | 0.6048 | 0.595 | | 69.81 | | SwiftVR | 2026-06-08 |
| SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training | ✓ Link | 22.46 | 0.274 | 0.600 | 0.4687 | | 66.01 | | SeedVR2-7B | 2025-06-05 |
| LiteVSR: Lightweight Adaptation of Frozen Diffusion Transformers for Video Super-Resolution | | 21.28 | 0.3842 | | 0.5741 | | 68.67 | 0.5984 | LiteVSR (Wan2.2-5B, 5-step full-rank adapter) (extra data) | 2026-06-08 |
| SeedVR: Seeding Infinity in Diffusion Transformer Towards Generic Video Restoration | ✓ Link | 21.15 | 0.298 | 0.554 | 0.635 | | 67.45 | 0.12788 | SeedVR-3B (extra data) | 2025-01-02 |
| SeedVR2: One-Step Video Restoration via Diffusion Adversarial Post-Training | ✓ Link | | | | | 1.376 | | 0.1334 | SeedVR2-3B | 2025-06-05 |
| SparkVSR: Interactive Video Super-Resolution via Sparse Keyframe Propagation | ✓ Link | | | | 0.6346 | | 69.1 | | SparkVSR (PiSA-SR reference) | 2026-03-17 |