# Detailed Caption on KVBench Detailed Caption, zh Source: [Detailed Caption on KVBench Detailed Caption, zh]() Collection index: [llms.txt]() Data updated: 2026-04-27T07:09:41.012Z Evaluation of Detailed Captioning performance on the Chinese version of the KVBench dataset, reporting overall and subject-specific scores across categories like Biology, Chemistry, and Math. 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 | Detailed Caption | | Dataset (including any stated split) | KVBench Detailed Caption, zh | | Metric | Overall Score | | Unit | SCALAR | | 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 - [Overall Score](): Higher is better; unit: SCALAR. - [Biology Score](): Higher is better; unit: SCALAR. - [Chemistry Score](): Higher is better; unit: SCALAR. - [Geography Score](): Higher is better; unit: SCALAR. - [History Score](): Higher is better; unit: SCALAR. - [Math Score](): Higher is better; unit: SCALAR. - [Physics Score](): Higher is better; unit: SCALAR. ## Recorded results | Method | Overall Score | Result date | Report paper | | --- | --- | --- | --- | | gemini-3-pro-image | 61.17 | Not recorded | 69eeed9955a3c32af89f97fd | | GPT-Image | 45.74 | Not recorded | 69eeed9955a3c32af89f97fd | | Seedream-4.0 | 44.17 | Not recorded | 69eeed9955a3c32af89f97fd | | gemini-2.5-flash | 27.37 | Not recorded | 69eeed9955a3c32af89f97fd | | Qwen-Image | 25.28 | Not recorded | 69eeed9955a3c32af89f97fd | | Z-Image | 25.01 | Not recorded | 69eeed9955a3c32af89f97fd | | FLUX.2-dev | 24.55 | Not recorded | 69eeed9955a3c32af89f97fd | | FLUX.2-max | 14.88 | Not recorded | 69eeed9955a3c32af89f97fd | | HiDream | 9.47 | Not recorded | 69eeed9955a3c32af89f97fd | | BAGEL-7B-MoT | 6.87 | Not recorded | 69eeed9955a3c32af89f97fd | | OmniGen2 | 6.55 | Not recorded | 69eeed9955a3c32af89f97fd | | FLUX.1-Krea-dev | 5.76 | Not recorded | 69eeed9955a3c32af89f97fd | | Janus-Pro | 5.75 | Not recorded | 69eeed9955a3c32af89f97fd | | SD3.5-Large | 5.71 | Not recorded | 69eeed9955a3c32af89f97fd | | FLUX.1-dev | 4.75 | Not recorded | 69eeed9955a3c32af89f97fd | | Show-o2 | 4.1 | Not recorded | 69eeed9955a3c32af89f97fd | ### gemini-3-pro-image Overall Score: 61.17. Result ID: 69ef0a9a47f34ce924d4bd14. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### GPT-Image Overall Score: 45.74. Result ID: 69ef0a9a47f34ce924d4bd13. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### Seedream-4.0 Overall Score: 44.17. Result ID: 69ef0a9a47f34ce924d4bd12. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### gemini-2.5-flash Overall Score: 27.37. Result ID: 69ef0a9a47f34ce924d4bd11. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### Qwen-Image Overall Score: 25.28. Result ID: 69ef0a9a47f34ce924d4bd1e. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### Z-Image Overall Score: 25.01. Result ID: 69ef0a9a47f34ce924d4bd1d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### FLUX.2-dev Overall Score: 24.55. Result ID: 69ef0a9a47f34ce924d4bd1c. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### FLUX.2-max Overall Score: 14.88. Result ID: 69ef0a9a47f34ce924d4bd10. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### HiDream Overall Score: 9.47. Result ID: 69ef0a9a47f34ce924d4bd0f. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Source Category":"Close-source Models"}. This record does not assert independent reproduction. ### BAGEL-7B-MoT Overall Score: 6.87. Result ID: 69ef0a9a47f34ce924d4bd1b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### OmniGen2 Overall Score: 6.55. Result ID: 69ef0a9a47f34ce924d4bd1a. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### FLUX.1-Krea-dev Overall Score: 5.76. Result ID: 69ef0a9a47f34ce924d4bd19. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### Janus-Pro Overall Score: 5.75. Result ID: 69ef0a9a47f34ce924d4bd18. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### SD3.5-Large Overall Score: 5.71. Result ID: 69ef0a9a47f34ce924d4bd17. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### FLUX.1-dev Overall Score: 4.75. Result ID: 69ef0a9a47f34ce924d4bd16. [Reporting paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. ### Show-o2 Overall Score: 4.1. Result ID: 69ef0a9a47f34ce924d4bd15. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Source Category":"Open-source Models"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.