| Qwen3.8-Max: A New Bar for Coding and Cowork | ✓ Link | 88.7 | Qwen3.8-Max | 2026-08-03 |
| A new era of intelligence with Gemini 3 | | 87.6 | Gemini 3 Pro | 2025-11-18 |
| dots3-note Preview: A Small but Mighty Step Toward Long-Horizon Agency in Real Life | ✓ Link | 86.78 | dots3-note Preview | 2026-08-13 |
| Kimi K2.5: Visual Agentic Intelligence | ✓ Link | 86.6 | Kimi K2.5 | 2026-02-02 |
| Introducing GPT-5.2 | | 85.9 | GPT-5.2 | 2025-12-11 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 84.7 | Qwen3.5-397B-A17B | 2026-02-16 |
| Qwen3.6 | ✓ Link | 84.4 | Qwen3.6-27B | 2026-04-21 |
| Qwen3.6 | ✓ Link | 83.7 | Qwen3.6-35B-A3B | 2026-04-21 |
| SiLVR: A Simple Language-based Video Reasoning Framework | ✓ Link | 82.7 | SiLVR | 2025-05-30 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 82.3 | Qwen3.5-27B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 82.0 | Qwen3.5-122B-A10B | 2026-02-16 |
| ERNIE 5.0 Technical Report | | 81.11 | ERNIE 5.0 | 2026-02-04 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 80.4 | Qwen3.5-35B-A3B | 2026-02-16 |
| Qwen3-VL Technical Report | ✓ Link | 80.0 | Qwen3-VL-235B-A22B-Thinking | 2025-11-26 |
| HopChain: Multi-Hop Data Synthesis for Generalizable Vision-Language Reasoning | | 80 | Qwen3.5-397B-A17B (RLVR w/ Multi-Hop) | 2026-03-17 |
| Kwai Keye-VL-2.0 Technical Report | ✓ Link | 80.0 | Keye-VL-2.0-30B-A3B | 2026-06-09 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 78.9 | Qwen3.5-9B | 2026-02-16 |
| GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning | ✓ Link | 74.7 | GLM-4.6V | 2025-07-01 |
| OmniReasoning: Pushing the Limits of Audio-Visual Joint Reasoning | ✓ Link | 74.4 | OmniReasoning-30B-A3B (extra data) | 2026-09-30 |
| CyberV: Cybernetics for Test-time Scaling in Video Understanding | ✓ Link | 74.3 | Qwen2.5-VL-72B + CyberV | 2025-06-09 |
| Graph-to-Frame RAG: Visual-Space Knowledge Fusion for Training-Free and Auditable Video Reasoning | | 71.2 | InternVL3.5-14B + G2F-RAG | 2026-04-06 |
| Valley3: Scaling Omni Foundation Models for E-commerce | | 68.7 | Valley3 32B Instruct | 2026-05-02 |
| SAGE: Training Smart Any-Horizon Agents for Long Video Reasoning with Reinforcement Learning | ✓ Link | 68.1 | SAGE-Flash | 2025-12-15 |
| LongCat-Flash-Omni Technical Report | ✓ Link | 67.5 | LongCat-Flash-Omni Instruct | 2025-10-31 |
| OneThinker: All-in-one Reasoning Model for Image and Video | ✓ Link | 66.2 | OneThinker-8B | 2025-12-02 |
| Kwai Keye-VL 1.5 Technical Report | ✓ Link | 66.0 | Keye-VL-1.5-8B | 2025-09-01 |
| Kimi-VL Technical Report | ✓ Link | 65.2 | Kimi-VL-A3B-Thinking-2506 | 2025-04-10 |
| Qwen2.5-VL Technical Report | ✓ Link | 60.2 | Qwen2.5-VL-72B | 2025-02-19 |
| ResAdapt: Adaptive Resolution for Efficient Multimodal Reasoning | ✓ Link | 59.6 | Qwen3-VL-8B + ResAdapt (128 frames, 22.9% retention) | 2026-03-30 |
| Kwai Keye-VL Technical Report | ✓ Link | 57.6 | Keye-VL-8B | 2025-07-02 |
| VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding | ✓ Link | 57.4 | VideoChat3-4B | 2026-07-16 |
| VIDEOP2R: Video Understanding from Perception to Reasoning | | 55.0 | VideoP2R | 2025-11-14 |
| VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting | ✓ Link | 54.4 | VisionCoach-7B | 2026-03-15 |
| Video-R1: Reinforcing Video Reasoning in MLLMs | ✓ Link | 52.7 | Video-R1-7B (10k RL steps, 32 frames) | 2025-03-27 |
| Kimi-VL Technical Report | ✓ Link | 52.6 | Kimi-VL-A3B-Instruct | 2025-04-10 |
| Demo-ICL: In-Context Learning for Procedural Video Knowledge Acquisition | ✓ Link | 52.6 | Demo-ICL (7B) | 2026-02-09 |
| From Pixels to Words -- Towards Native One-Vision Models at Scale | ✓ Link | 51.6 | NEO-ov 9B | 2026-05-27 |
| Reinforced Attention Learning | | 49.2 | RAL-zero | 2026-02-04 |
| Qwen2.5-VL Technical Report | ✓ Link | 47.4 | Qwen2.5-VL-7B | 2025-02-19 |
| EventLens: Event-Structure Reinforcement Learning for Video Understanding | ✓ Link | 39.6 | EventLens-8B | 2026-10-12 |
| CoPE-VideoLM: Codec Primitives For Efficient Video Language Models | | 38.2 | CoPE 7B | 2026-02-13 |