# Video Understanding on MLVU (MLVU, Avg) Source: [Video Understanding on MLVU (MLVU, Avg)]() Collection index: [llms.txt]() Data updated: 2026-03-06T08:51:53.463Z Evaluation of Video Understanding models on the MLVU dataset, reporting the MLVU score and the Average metric. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Video Understanding | | Dataset (including any stated split) | MLVU | | Metric | MLVU Score | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 16 of 16 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [MLVU Score](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. ## Recorded results | Method | MLVU Score | Result date | Report paper | | --- | --- | --- | --- | | FLoC | 73.83 | Not recorded | 690988763ef7ccd74e5e5a3e | | DivPrune | 73.37 | Not recorded | 690988763ef7ccd74e5e5a3e | | TS-LLaVA | 73.09 | Not recorded | 690988763ef7ccd74e5e5a3e | | FLoC | 72.63 | Not recorded | 690988763ef7ccd74e5e5a3e | | Random | 72.49 | Not recorded | 690988763ef7ccd74e5e5a3e | | TS-LLaVA | 72.4 | Not recorded | 690988763ef7ccd74e5e5a3e | | DyCoke | 72.26 | Not recorded | 690988763ef7ccd74e5e5a3e | | LongVU | 72.22 | Not recorded | 690988763ef7ccd74e5e5a3e | | - | 71.57 | Not recorded | 690988763ef7ccd74e5e5a3e | | DyCoke | 71.3 | Not recorded | 690988763ef7ccd74e5e5a3e | | LongVU | 71.02 | Not recorded | 690988763ef7ccd74e5e5a3e | | Random | 70.52 | Not recorded | 690988763ef7ccd74e5e5a3e | | - | 70.31 | Not recorded | 690988763ef7ccd74e5e5a3e | | DivPrune | 70.19 | Not recorded | 690988763ef7ccd74e5e5a3e | | PruneVid | 70.19 | Not recorded | 690988763ef7ccd74e5e5a3e | | PruneVid | 68.63 | Not recorded | 690988763ef7ccd74e5e5a3e | ### FLoC MLVU Score: 73.83. Result ID: 69aa95a26dbb7075b954733d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### DivPrune MLVU Score: 73.37. Result ID: 69aa95a26dbb7075b9547339. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### TS-LLaVA MLVU Score: 73.09. Result ID: 69aa95a26dbb7075b9547337. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### FLoC MLVU Score: 72.63. Result ID: 69aa95a26dbb7075b9547335. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### Random MLVU Score: 72.49. Result ID: 69aa95a26dbb7075b954733a. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### TS-LLaVA MLVU Score: 72.4. Result ID: 69aa95a26dbb7075b954732f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### DyCoke MLVU Score: 72.26. Result ID: 69aa95a26dbb7075b954733b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### LongVU MLVU Score: 72.22. Result ID: 69aa95a26dbb7075b9547338. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### - MLVU Score: 71.57. Result ID: 69aa95a26dbb7075b9547336. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"768"}. This record does not assert independent reproduction. ### DyCoke MLVU Score: 71.3. Result ID: 69aa95a26dbb7075b9547333. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### LongVU MLVU Score: 71.02. Result ID: 69aa95a26dbb7075b9547330. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### Random MLVU Score: 70.52. Result ID: 69aa95a26dbb7075b9547332. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### - MLVU Score: 70.31. Result ID: 69aa95a26dbb7075b954732e. [Reporting paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"768"}. This record does not assert independent reproduction. ### DivPrune MLVU Score: 70.19. Result ID: 69aa95a26dbb7075b9547331. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. ### PruneVid MLVU Score: 70.19. Result ID: 69aa95a26dbb7075b954733c. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-32B","Max Frames":"7200"}. This record does not assert independent reproduction. ### PruneVid MLVU Score: 68.63. Result ID: 69aa95a26dbb7075b9547334. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model":"Qwen2.5-VL-7B","Max Frames":"7200"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.