# Long Video Question Answering on Video-MME w/o subtitles Source: [Long Video Question Answering on Video-MME w/o subtitles]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:31:56.679Z Evaluation of model performance on the Video-MME dataset subset excluding subtitles, specifically focusing on answering long questions via Video Question Answering. 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 | Long Video Question Answering | | Dataset (including any stated split) | Video-MME w/o subtitles | | Metric | Accuracy | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 14 of 14 | | 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 - [Accuracy](): Higher is better; unit: SCALAR. ## Recorded results | Method | Accuracy | Result date | Report paper | | --- | --- | --- | --- | | GPT-5 | 0.818 | Not recorded | 6938fe806cfd488761a53c0c | | Gemini-1.5-Pro | 0.75 | Not recorded | 6938fe806cfd488761a53c0c | | GPT-4o | 0.719 | Not recorded | 6938fe806cfd488761a53c0c | | InternVL3.5 | 0.66 | Not recorded | 6938fe806cfd488761a53c0c | | AKS | 0.653 | Not recorded | 6938fe806cfd488761a53c0c | | Qwen2.5-VL | 0.651 | Not recorded | 6938fe806cfd488761a53c0c | | LLaVA-Video + OneClip-RAG | 0.649 | Not recorded | 6938fe806cfd488761a53c0c | | ByteVideoLLM | 0.646 | Not recorded | 6938fe806cfd488761a53c0c | | NVILA | 0.642 | Not recorded | 6938fe806cfd488761a53c0c | | LLaVA-Video | 0.633 | Not recorded | 6938fe806cfd488761a53c0c | | LongVU | 0.606 | Not recorded | 6938fe806cfd488761a53c0c | | mPLUG-Owl3 | 0.593 | Not recorded | 6938fe806cfd488761a53c0c | | Video-XL | 0.555 | Not recorded | 6938fe806cfd488761a53c0c | | LongVA | 0.526 | Not recorded | 6938fe806cfd488761a53c0c | ### GPT-5 Accuracy: 0.818. Result ID: 69986f685698665ec4cf6066. [Reporting paper]() Recorded evaluation configuration: {"LLM":"N/A","# Frames":"N/A"}. This record does not assert independent reproduction. ### Gemini-1.5-Pro Accuracy: 0.75. Result ID: 69986f685698665ec4cf6068. [Reporting paper]() Recorded evaluation configuration: {"LLM":"N/A","# Frames":"N/A"}. This record does not assert independent reproduction. ### GPT-4o Accuracy: 0.719. Result ID: 69986f685698665ec4cf6067. [Reporting paper]() Recorded evaluation configuration: {"LLM":"N/A","# Frames":"N/A"}. This record does not assert independent reproduction. ### InternVL3.5 Accuracy: 0.66. Result ID: 69986f685698665ec4cf606e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"8B","# Frames":"64"}. This record does not assert independent reproduction. ### AKS Accuracy: 0.653. Result ID: 69986f685698665ec4cf606a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"64"}. This record does not assert independent reproduction. ### Qwen2.5-VL Accuracy: 0.651. Result ID: 69986f685698665ec4cf606d. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"32"}. This record does not assert independent reproduction. ### LLaVA-Video + OneClip-RAG Accuracy: 0.649. Result ID: 69986f685698665ec4cf6073. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"64"}. This record does not assert independent reproduction. ### ByteVideoLLM Accuracy: 0.646. Result ID: 69986f685698665ec4cf6071. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"14B","# Frames":"256"}. This record does not assert independent reproduction. ### NVILA Accuracy: 0.642. Result ID: 69986f685698665ec4cf6070. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"8B","# Frames":"256"}. This record does not assert independent reproduction. ### LLaVA-Video Accuracy: 0.633. Result ID: 69986f685698665ec4cf6072. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"64"}. This record does not assert independent reproduction. ### LongVU Accuracy: 0.606. Result ID: 69986f685698665ec4cf606f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"1fps"}. This record does not assert independent reproduction. ### mPLUG-Owl3 Accuracy: 0.593. Result ID: 69986f685698665ec4cf606c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"16"}. This record does not assert independent reproduction. ### Video-XL Accuracy: 0.555. Result ID: 69986f685698665ec4cf6069. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"128"}. This record does not assert independent reproduction. ### LongVA Accuracy: 0.526. Result ID: 69986f685698665ec4cf606b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"LLM":"7B","# Frames":"256"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.