ComfyUI Node
ROCm Flux Benchmark
Comprehensive Flux workflow benchmark for AMD GPUs
ROCm Flux Benchmark
- model
- vae
- clip
- BENCHMARK_RESULTS
- PERFORMANCE_CHART
- OPTIMIZATION_RECOMMENDATIONS
- MEMORY_ANALYSIS
◄test_resolutions256x320,512x512,1024x1024►
◄test_steps20►
◄test_cfg_values1.0,3.5,8.0►
◄test_hipblastrue►
◄generate_reporttrue►
CategoryROCm Ninodes/Benchmark
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| model | MODEL | The model to benchmark | |
| vae | VAE | The VAE model | |
| clip | CLIP | The CLIP model | |
| test_resolutions | STRING | 256x320,512x512,1024x1024 | Comma-separated resolutions to test (WxH) |
| test_steps | INT | 201–100 | Number of sampling steps for testing |
| test_cfg_values | STRING | 1.0,3.5,8.0 | Comma-separated CFG values to test |
| test_hipblas | BOOLEAN | true | Test HIPBlas optimizations |
| generate_report | BOOLEAN | true | Generate detailed optimization report |
Outputs (4)
| Name | Type | Description |
|---|---|---|
| BENCHMARK_RESULTS | STRING | — |
| PERFORMANCE_CHART | STRING | — |
| OPTIMIZATION_RECOMMENDATIONS | STRING | — |
| MEMORY_ANALYSIS | STRING | — |