# Text-to-Video on MS-COCO random 1,000 prompts 2017 (test) Source: [Text-to-Video on MS-COCO random 1,000 prompts 2017 (test)]() Collection index: [llms.txt]() Data updated: 2026-05-28T09:22:05.029Z Evaluation of Text-to-Video models using a specific subset of 1,000 random prompts from the MS-COCO 2017 test set, assessed using metrics like Textual Faithfulness, Frame Consistency, and generation Speedup. 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 | Text-to-Video | | Dataset (including any stated split) | MS-COCO random 1,000 prompts 2017 (test) | | Metric | Textual Faithfulness | | 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 - [Textual Faithfulness](): Higher is better; unit: SCALAR. - [Frame Consistency](): Higher is better; unit: PERCENT. - [Speedup (x)](): Higher is better; unit: SCALAR. ## Recorded results | Method | Textual Faithfulness | Result date | Report paper | | --- | --- | --- | --- | | CogVideoX 2B | 0.3918 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.3775 | Not recorded | 68fa7af61e209131cea86010 | | Animated-Diff | 0.3662 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.3465 | Not recorded | 68fa7af61e209131cea86010 | | Animated-Diff | 0.305 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.2939 | Not recorded | 68fa7af61e209131cea86010 | | Animated-Diff | 0.2439 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.2426 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.232 | Not recorded | 68fa7af61e209131cea86010 | | CogVideoX 2B | 0.2302 | Not recorded | 68fa7af61e209131cea86010 | | CogVideoX 2B (int8) | 0.2113 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | 0.201 | Not recorded | 68fa7af61e209131cea86010 | | CogVideoX 2B | -0.1441 | Not recorded | 68fa7af61e209131cea86010 | | LTC-Accel | -0.1673 | Not recorded | 68fa7af61e209131cea86010 | ### CogVideoX 2B Textual Faithfulness: 0.3918. Result ID: 6a180866fbc373592c120cc8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"40"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.3775. Result ID: 6a180866fbc373592c120cc9. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"CogVideoX 2B","Steps":"26"}. This record does not assert independent reproduction. ### Animated-Diff Textual Faithfulness: 0.3662. Result ID: 6a180866fbc373592c120cc2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"30"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.3465. Result ID: 6a180866fbc373592c120cc3. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Animated-Diff","Steps":"19"}. This record does not assert independent reproduction. ### Animated-Diff Textual Faithfulness: 0.305. Result ID: 6a180866fbc373592c120cc0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"20"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.2939. Result ID: 6a180866fbc373592c120cc1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Animated-Diff","Steps":"13"}. This record does not assert independent reproduction. ### Animated-Diff Textual Faithfulness: 0.2439. Result ID: 6a180866fbc373592c120cbe. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"10"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.2426. Result ID: 6a180866fbc373592c120cbf. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"Animated-Diff","Steps":"7"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.232. Result ID: 6a180866fbc373592c120cc7. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"CogVideoX 2B","Steps":"19"}. This record does not assert independent reproduction. ### CogVideoX 2B Textual Faithfulness: 0.2302. Result ID: 6a180866fbc373592c120cc6. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"30"}. This record does not assert independent reproduction. ### CogVideoX 2B (int8) Textual Faithfulness: 0.2113. Result ID: 6a180866fbc373592c120cca. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"40","Precision":"int8"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: 0.201. Result ID: 6a180866fbc373592c120ccb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"CogVideoX 2B","Steps":"26","Precision":"int8"}. This record does not assert independent reproduction. ### CogVideoX 2B Textual Faithfulness: -0.1441. Result ID: 6a180866fbc373592c120cc4. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Sampler":"Original","Steps":"20"}. This record does not assert independent reproduction. ### LTC-Accel Textual Faithfulness: -0.1673. Result ID: 6a180866fbc373592c120cc5. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Base Model":"CogVideoX 2B","Steps":"14"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.