| A Simple Baseline for Streaming Video Understanding | | 81.4 | 54.0 | | SimpleStream (Qwen3-VL-8B, 4 frames) | 2026-04-02 |
| A Simple Baseline for Streaming Video Understanding | | 79.9 | 54.9 | | SimpleStream (Qwen3-VL-8B, 8 frames) | 2026-04-02 |
| A Simple Baseline for Streaming Video Understanding | | 79.3 | 53.5 | | SimpleStream (Qwen3-VL-8B, 2 frames) | 2026-04-02 |
| A Simple Baseline for Streaming Video Understanding | | 78.4 | 51.9 | | SimpleStream (Qwen2.5-VL-7B, 4 frames) | 2026-04-02 |
| A Simple Baseline for Streaming Video Understanding | | 76.6 | 50.8 | | SimpleStream (Qwen2.5-VL-7B, 8 frames) | 2026-04-02 |
| A Simple Baseline for Streaming Video Understanding | | 75.0 | 49.5 | | SimpleStream (Qwen2.5-VL-7B, 2 frames) | 2026-04-02 |
| An Efficient Streaming Video Understanding Framework with Agentic Control | | 71.89 | 51.27 | 50.60 | R3-Streaming (7B / 4B-Thinking) | 2026-05-18 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 69.32 | 62.54 | 57.15 | Gemini 1.5 Pro | 2025-01-09 |
| StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video | | 67.9 | 46.0 | | StreamPro-SFT (4B) | 2026-05-11 |
| FluxMem: Adaptive Hierarchical Memory for Streaming Video Understanding | | 67.2 | | | FluxMem-7B | 2026-03-02 |
| Thinking in Streaming Video | | 67.03 | 52.30 | | ThinkStream-3B | 2026-03-13 |
| StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video | | 66.9 | 45.2 | | StreamPro-GRPO (4B) | 2026-05-11 |
| ViSpeak: Visual Instruction Feedback in Streaming Videos | | 66.28 | 57.52 | 54.25 | ViSpeak-7B | 2025-03-17 |
| Streaming Video Instruction Tuning | | 65.98 | 46.10 | 54.77 | Streamo-7B | 2025-12-24 |
| An Efficient Streaming Video Understanding Framework with Agentic Control | | 65.12 | 44.64 | 52.44 | R3-Streaming (3B / 7B-Instruct) | 2026-05-18 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 64.46 | 60.75 | 53.40 | GPT-4o | 2025-01-09 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 64.02 | 43.71 | 50.50 | LLaVA-OneVision-7B | 2025-01-09 |
| An Efficient Streaming Video Understanding Framework with Agentic Control | | 63.58 | 50.23 | 55.83 | R3-Streaming (3B / 4B-Thinking) | 2026-05-18 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 63.52 | 40.4 | 54.82 | LLaVA-Video-7B | 2025-01-09 |
| StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video | | 62.5 | 34.9 | | StreamPro-GRPO (3B) | 2026-05-11 |
| StreamPro: From Reactive Perception to Proactive Decision-Making in Streaming Video | | 62.0 | 40.0 | | StreamPro-SFT (3B) | 2026-05-11 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 61.92 | 56.95 | 49.30 | Qwen2-VL-72B | 2025-01-09 |
| TimeChat-Online: 80% Visual Tokens are Naturally Redundant in Streaming Videos | | 61.9 | 41.7 | 36.7 | TimeChat-Online-7B | 2025-04-24 |
| Streaming Video Instruction Tuning | | 61.51 | 41.76 | 53.72 | Streamo-3B | 2025-12-24 |
| StreamAgent: Towards Anticipatory Agents for Streaming Video Understanding | | 61.3 | 41.7 | 45.4 | StreamAgent-7B | 2025-08-03 |
| Thinking in Streaming Video | | 61.20 | 52.02 | | StreamForest-7B | 2026-03-13 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 60.39 | 43.44 | 46.60 | InternVL2-8B | 2025-01-09 |
| A Simple Baseline for Streaming Video Understanding | | 59.9 | 44.7 | | Qwen2.5-VL-7B | 2026-04-02 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 57.61 | 35.01 | 47.50 | LongVU-7B | 2025-01-09 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 55.98 | 46.46 | 48.74 | Qwen2-VL-7B | 2025-01-09 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 54.55 | 36.06 | 34.72 | Dispider-7B | 2025-01-09 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 28.37 | 27.38 | 45.09 | Flash-VStream-7B | 2025-01-09 |
| OVO-Bench: How Far is Your Video-LLMs from Real-World Online Video Understanding? | | 20.79 | 17.73 | | VideoLLM-online-8B | 2025-01-09 |