| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 83.9 | Qwen3.5-122B-A10B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 83.7 | Qwen3.5-397B-A17B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 82.8 | Qwen3.5-27B | 2026-02-16 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 82.5 | Qwen3.5-35B-A3B | 2026-02-16 |
| Qwen3.5-Omni Technical Report | | 81.9 | Qwen3.5-Omni-Plus | 2026-04-21 |
| VideoJev: System One Model for Video Understanding and Interaction | ✓ Link | 81.8 | VideoJev-35B-A3B | 2026-10-13 |
| Qwen3-VL Technical Report | ✓ Link | 79.2 | Qwen3-VL-235B-A22B-Instruct | 2025-11-26 |
| Qwen3.5: Towards Native Multimodal Agents | ✓ Link | 78.4 | Qwen3.5-9B | 2026-02-16 |
| Kwai Keye-VL-2.0 Technical Report | ✓ Link | 78.3 | Keye-VL-2.0-30B-A3B | 2026-06-09 |
| Video-RAG: Visually-aligned Retrieval-Augmented Long Video Comprehension | ✓ Link | 77.4 | Video-RAG (based on LLaVA-Video) | 2024-11-20 |
| Annotations as Rollouts: Efficient and Scalable Reinforcement Learning for Video MLLMs | ✓ Link | 76.7 | Video-ORA-9B | 2026-08-20 |
| TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live Streaming | ✓ Link | 75.6 | TLive-Omni (9B) | 2026-08-21 |
| VideoJev: System One Model for Video Understanding and Interaction | ✓ Link | 75.0 | VideoJev-4B | 2026-10-13 |
| GLM-4.1V-Thinking: Towards Versatile Multimodal Reasoning with Scalable Reinforcement Learning | ✓ Link | 74.8 | GLM-4.6V | 2025-07-01 |
| Capek 0.5: An Execution-Centric Vision-Language Model for Embodied Intelligence | | 74.19 | Capek0.5-35B-A3B | 2026-08-07 |
| InternVideo3: Agentify Foundation Models with Multimodal Contextual Reasoning | ✓ Link | 73.8 | InternVideo3-8B | 2026-06-10 |
| Qwen2.5-VL Technical Report | ✓ Link | 73.3 | Qwen2.5-VL-72B | 2025-02-19 |
| Kwai Keye-VL 1.5 Technical Report | ✓ Link | 73.0 | Keye-VL-1.5-8B | 2025-09-01 |
| InternVL3: Exploring Advanced Training and Test-Time Recipes for Open-Source Multimodal Models | ✓ Link | 72.7 | InternVL3-78B | 2025-04-14 |
| Nemotron 3 Nano Omni: Efficient and Open Multimodal Intelligence | ✓ Link | 72.2 | Nemotron 3 Nano Omni (Reasoning) | 2026-04-27 |
| Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context | ✓ Link | 71.9 | Gemini 1.5 Pro | 2024-03-08 |
| LLaVA-OneVision-2: Towards Next-Generation Perceptual Intelligence | ✓ Link | 71.9 | LLaVA-OneVision-2-8B | 2026-05-25 |
| Less from More: Reinforcing Sparse Video Reasoning from Dense References | ✓ Link | 71.8 | SAVER-4B (extra data) | 2026-10-07 |
| VideoAuto-R1: Video Auto Reasoning via Thinking Once, Answering Twice | ✓ Link | 71.7 | VideoAuto-R1 (Qwen3-VL-8B) | 2026-01-08 |
| Qwen3-Omni Technical Report | ✓ Link | 71.4 | Qwen3-Omni-Flash-Instruct | 2025-09-22 |
| Penguin-VL: Exploring the Efficiency Limits of VLM with LLM-based Vision Encoders | ✓ Link | 71.4 | Penguin-VL (8B) | 2026-03-06 |
| TLive-Omni: An Omni-Modal Understanding Model for E-Commerce Live Streaming | ✓ Link | 71.3 | TLive-Omni (4B) | 2026-08-21 |
| Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution | ✓ Link | 71.2 | Qwen2-VL-72B | 2024-09-18 |
| Audio-Visual Flamingo: Open Audio-Visual Intelligence for Long and Complex Videos | | 70.7 | AVF-Instruct | 2026-07-17 |
| MiniCPM-o 4.5: Towards Real-Time Full-Duplex Omni-Modal Interaction | ✓ Link | 70.4 | MiniCPM-o 4.5-Instruct | 2026-04-30 |
| Enhancing Video Vision Language Model with Hippocampal Sensing | | 70.4 | HippoVLM-8B | 2026-06-03 |
| GPT-4o: Visual perception performance of multimodal large language models in piglet activity understanding | | 70.3 | GPT-4o | 2024-06-14 |
| VideoChat3: Fully Open Video MLLM for Efficient and Generalist Video Understanding | ✓ Link | 70.1 | VideoChat3-4B | 2026-07-16 |
| Molmo2: Open Weights and Data for Vision-Language Models with Video Understanding and Grounding | ✓ Link | 69.9 | Molmo2-8B | 2026-01-15 |
| PhysBrain 1.5: From Vision-Language Models to Physical Foundation Models | ✓ Link | 68.22 | PhysBrain 1.5-8B | 2026-09-14 |
| MOSS-VL Technical Report | ✓ Link | 68.1 | MOSS-VL-Instruct-0708 | 2026-08-15 |
| Kimi-VL Technical Report | ✓ Link | 67.8 | Kimi-VL-A3B-Instruct | 2025-04-10 |
| Insight-V++: Towards Advanced Long-Chain Visual Reasoning with Multimodal Large Language Models | | 67.8 | Insight-V++ (Self-Evolving, Qwen2.5-VL, 7B) | 2026-03-18 |
| Small Vision-Language Models are Smart Compressors for Long Video Understanding | ✓ Link | 67.8 | Tempo | 2026-04-09 |
| Kwai Keye-VL Technical Report | ✓ Link | 67.7 | Keye-VL-8B | 2025-07-02 |
| From Pixels to Words -- Towards Native One-Vision Models at Scale | ✓ Link | 67.4 | NEO-ov 9B | 2026-05-27 |
| VisReflect: Latent Visual Reflection for Fine-Grained Perception in Long Visual Context | ✓ Link | 67.4 | VisReflect-7B-Video | 2026-06-29 |
| Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context | ✓ Link | 66.3 | Gemini 1.5 Flash | 2024-03-08 |
| LLaVA-OneVision: Easy Visual Task Transfer | ✓ Link | 66.2 | LLaVA-OneVision-72B | 2024-08-06 |
| VideoLLaMA 3: Frontier Multimodal Foundation Models for Image and Video Understanding | ✓ Link | 66.2 | VideoLLaMA3 7B | 2025-01-22 |
| Reinforced Attention Learning | | 65.1 | RAL-zero | 2026-02-04 |
| AURA: Always-On Understanding and Real-Time Assistance via Video Streams | ✓ Link | 65.1 | AURA (8B) | 2026-04-05 |
| | 64.8 | LLaVA-OneVision (72B) | |
| Structured Causal Video Reasoning via Multi-Objective Alignment | | 64.7 | STEER-4B | 2026-04-06 |
| Watch Before You Answer: Learning from Visually Grounded Post-Training | ✓ Link | 63.4 | VidGround (64 frames) | 2026-04-06 |
| Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution | ✓ Link | 63.3 | Qwen2-VL-7B | 2024-09-18 |
| VisionCoach: Reinforcing Grounded Video Reasoning via Visual-Perception Prompting | ✓ Link | 63.3 | VisionCoach-7B | 2026-03-15 |
| PointThinker: Point-Incentivized Parallel Thinking for Multimodal Large Language Model | | 62.5 | PointThinker-7B (extra data) | 2026-06-03 |
| FlexiVideo: Variation-Aware Temporal Dynamics Modeling for Efficient Video Understanding | ✓ Link | 62.5 | FlexiVideo-3B (extra data) | 2026-06-03 |
| GPT-4o: Visual perception performance of multimodal large language models in piglet activity understanding | | 62.3 | GPT-4o mini | 2024-06-14 |
| Incentivizing Versatile Video Reasoning in MLLMs via Data-Efficient Reinforcement Learning | ✓ Link | 62.0 | VideoReasoner | 2026-06-03 |
| VILA: On Pre-training for Visual Language Models | ✓ Link | 61.4 | VILA-1.5 (34B) | 2023-12-12 |
| Dynin-Omni: Omnimodal Unified Large Diffusion Language Model | ✓ Link | 61.4 | Dynin-Omni (8B) | 2026-03-09 |
| VideoLLaMA 2: Advancing Spatial-Temporal Modeling and Audio Understanding in Video-LLMs | ✓ Link | 60.9 | VideoLLaMA2 (72B) | 2024-06-11 |
| OmniStream: Mastering Perception, Reconstruction and Action in Continuous Streams | ✓ Link | 60.7 | OmniStream-7B | 2026-03-12 |
| EVA: Efficient Reinforcement Learning for End-to-End Video Agent | ✓ Link | 59.9 | EVA (GRPO) | 2026-03-24 |
| TimeViper: A Hybrid Mamba-Transformer Vision-Language Model for Efficient Long Video Understanding | ✓ Link | 58.8 | TimeViper | 2025-11-20 |
| PixelUMM: Encoder-Free Unified Image and Video Understanding and Generation | ✓ Link | 57.33 | PixelUMM 8B MoT | 2026-09-29 |
| Qwen2-VL: Enhancing Vision-Language Model's Perception of the World at Any Resolution | ✓ Link | 55.6 | Qwen2-VL-2B | 2024-09-18 |
| Generation as Auxiliary Supervision: Enhancing Visual Understanding at Zero Inference Overhead via Decoupled Embedding Prediction | | 54.4 | GAS (4B) | 2026-08-12 |
| MoE-ViE: Mixture of Experts Vision Encoder for Efficient Image and Video Understanding | ✓ Link | 53.0 | MoE-ViE-H/14 + Llama 3.1 8B (42M alignment) | 2026-08-18 |
| SmolVLM: Redefining small and efficient multimodal models | ✓ Link | 52.1 | SmolVLM 2.2B | 2025-04-07 |
| TimeSuite: Improving MLLMs for Long Video Understanding via Grounded Tuning | ✓ Link | 46.3 | VideoChat-T (7B) | 2024-10-25 |