# Text-to-Image Generation on FLUX Source: [Text-to-Image Generation on FLUX]() Collection index: [llms.txt]() Data updated: 2026-03-24T14:15:36.048Z Evaluation of Text-to-Image generation models on the FLUX dataset, focusing on resource utilization metrics including Training Memory, Parameter Memory, and TFLOPs. 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-Image Generation | | Dataset (including any stated split) | FLUX | | Metric | Training Memory (GiB) | | Unit | SIZE | | Direction | Lower is better | | Results loaded | 11 of 11 | | 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 - [Training Memory (GiB)](): Lower is better; unit: SIZE. - [Parameter Memory (GiB)](): Lower is better; unit: SIZE. - [Optimizer State Memory (GiB)](): Lower is better; unit: SIZE. - [Other Memory (GiB)](): Lower is better; unit: SIZE. - [Training TFLOPs](): Lower is better; unit: SCALAR. - [Precompute TFLOPs](): Lower is better; unit: SCALAR. ## Recorded results | Method | Training Memory (GiB) | Result date | Report paper | | --- | --- | --- | --- | | DiT-BlockSkip | 10.42 | Not recorded | 69c21a690636e5bf4afc4697 | | HollowedNet | 12.23 | Not recorded | 69c21a690636e5bf4afc4697 | | DiT-BlockSkip | 15.61 | Not recorded | 69c21a690636e5bf4afc4697 | | HollowedNet | 18.4 | Not recorded | 69c21a690636e5bf4afc4697 | | DiT-BlockSkip | 20.78 | Not recorded | 69c21a690636e5bf4afc4697 | | HollowedNet | 24.11 | Not recorded | 69c21a690636e5bf4afc4697 | | LoRA-FA | 32.48 | Not recorded | 69c21a690636e5bf4afc4697 | | LISA | 35.98 | Not recorded | 69c21a690636e5bf4afc4697 | | LoRA | 35.99 | Not recorded | 69c21a690636e5bf4afc4697 | | Textual Inversion | 42.15 | Not recorded | 69c21a690636e5bf4afc4697 | | DreamBooth | 68.96 | Not recorded | 69c21a690636e5bf4afc4697 | ### DiT-BlockSkip Training Memory (GiB): 10.42. Result ID: 69c29c7c47f34ce924ab30f2. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"50 %","Training Resolution":"256×256","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### HollowedNet Training Memory (GiB): 12.23. Result ID: 69c29c7c47f34ce924ab30ef. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"50 %","Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### DiT-BlockSkip Training Memory (GiB): 15.61. Result ID: 69c29c7c47f34ce924ab30f1. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"40 %","Training Resolution":"256×256","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### HollowedNet Training Memory (GiB): 18.4. Result ID: 69c29c7c47f34ce924ab30ee. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"40 %","Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### DiT-BlockSkip Training Memory (GiB): 20.78. Result ID: 69c29c7c47f34ce924ab30f0. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"30 %","Training Resolution":"256×256","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### HollowedNet Training Memory (GiB): 24.11. Result ID: 69c29c7c47f34ce924ab30ed. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Skip Ratio":"30 %","Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### LoRA-FA Training Memory (GiB): 32.48. Result ID: 69c29c7c47f34ce924ab30ec. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### LISA Training Memory (GiB): 35.98. Result ID: 69c29c7c47f34ce924ab30eb. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### LoRA Training Memory (GiB): 35.99. Result ID: 69c29c7c47f34ce924ab30ea. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### Textual Inversion Training Memory (GiB): 42.15. Result ID: 69c29c7c47f34ce924ab30e8. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. ### DreamBooth Training Memory (GiB): 68.96. Result ID: 69c29c7c47f34ce924ab30e9. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Training Resolution":"512×512","Inference resolution":"1024 × 1024"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.