Benchmark
LongVideoBench
LongVideoBench is a question-answering benchmark featuring video-language interleaved inputs up to an hour long. It includes 3,763 varying-length…
- Modality
- multimodal
- Categories
- long_context, multimodal, vision
- Openness
- unknown
- Source
- llm_stats
- Reported scores
- 4
Reported scores
llm_stats
| Model | Organization | Reported value | Reported | Evidence |
|---|---|---|---|---|
| Kimi K2.5 | Moonshot AI | 0.798 | 2026-01-27 | evidence |
| Qwen2.5 VL 7B Instruct | Qwen | 0.547 | 2025-01-26 | evidence |
| Seed 2.1 Pro | ByteDance | 0.806 | 2026-06-24 | evidence |
| Seed 2.1 Turbo | ByteDance | 0.806 | 2026-06-24 | evidence |
Scores are partitioned by the source that reported them and are never merged into a single cross-source ranking, because the sources measure different things and say so.