Mixed DiT Serving on Uniform workload mix 12 req/min, SLO Scale = 1.0x FLUX on H100
88SLO Attainment Ratio (SAR)TetriServe
Evaluation Results
| Method | Links | |
|---|---|---|
| TetriServeNirvana Integration=true, Hardware=H100, Model=FLUX2025.10 | 88 | |
| RSSPNirvana Integration=true, Hardware=H100, Model=FLUX2025.10 | 77 | |
| TetriServeNirvana Integration=false, Hardware=H100, Model=FLUX2025.10 | 42 | |
| RSSPNirvana Integration=false, Hardware=H100, Model=FLUX2025.10 | 32 |