ComfyUI Node
⚗ Alchemica · Schedule
A ComfyUI node in ⚗ AlchemicaMente/sampling with 12 inputs and 1 output.
⚗ Alchemica · Schedule
- opts
- ALCHEMICA_OPTS
◄scheduleralchemica_duale►
◄steps8►
◄duale_alto1.00►
◄duale_ampiezza_alto0.10►
◄duale_basso8.0►
◄duale_ampiezza_basso0.08►
◄tarato_esponente0.50►
◄tarato_miscela0.15►
◄shift_override0.00►
◄flux_q1.00►
◄plunge_sigma0.00►
Category⚗ AlchemicaMente/sampling
Inputs (12)
| Name | Type | Default | Description |
|---|---|---|---|
| scheduler | COMBO | alchemica_duale | 13 options: alchemica_duale, alchemica_tarato, flux_shift, linear_flow, comfy:simple, comfy:sgm_uniform, +7 |
| steps | INT | 81–200 | Total steps (single stage) |
| duale_alto | FLOAT | 1.000–20 | alchemica_duale: extra steps on composition (high sigma) |
| duale_ampiezza_alto | FLOAT | 0.100.01–1 | alchemica_duale: how far the high lobe reaches down |
| duale_basso | FLOAT | 8.00–40 | alchemica_duale: extra steps on the final detail (low sigma) |
| duale_ampiezza_basso | FLOAT | 0.080.01–1 | alchemica_duale: how far the low lobe reaches up |
| tarato_esponente | FLOAT | 0.500–1.5 | alchemica_tarato: 0.5 = theoretical optimum for euler. Higher = more steps where the model changes most |
| tarato_miscela | FLOAT | 0.150–1 | alchemica_tarato: share of steps spread uniformly |
| shift_override | FLOAT | 0.000–5 | Shift mu (0 = the model's own, 1.15 for Krea 2) |
| flux_q | FLOAT | 1.000.3–3 | flux_shift: >1 more steps at the top, <1 at the bottom |
| plunge_sigma | FLOAT | 0.000–0.95 | 0 = off. Otherwise structure stops here and the last step jumps to 0 |
| optsopt | ALCHEMICA_OPTS | Chain from another Alchemica options node |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| ALCHEMICA_OPTS | ALCHEMICA_OPTS | — |