Rapid Scheduler Combo
Rapid Scheduler Combo
- sampler
- scheduler
Rapid Scheduler Combo is the laziest node in this pack, and that's the point. It exists because the rapidfire workflow - the model-chaining samplers this pack is built around - has sampler and scheduler dropdowns scattered across every node, and keeping them in sync across a graph gets old fast. This node gives you one little box with two dropdowns, and its entire job is to hand those two strings to whatever sampler node actually does the work.
It doesn't call anything, it doesn't sample anything, and it has no opinion about your prompt. You pick a sampler from the 44-item list and a scheduler from the 9-item list, and it passes them straight through as outputs.
How it works
Mechanically it's a passthrough. The sampler input is a dropdown over ComfyUI's full KSampler.SAMPLERS list (euler, euler_ancestral, dpmpp_2m, dpmpp_sde, lcm, the whole family), and scheduler is the standard KSampler.SCHEDULERS list (simple, sgm_uniform, karras, exponential, ddim_uniform, beta, normal, linear_quadratic, kl_optimal). Both outputs are typed strings, so the thing they're wired into has to accept sampler/scheduler values as wires - that's exactly what MultiModelAdvancedKsampler and the other MultiModel nodes in this pack do, and what several third-party advanced sampler packs accept.
What you actually set
Just two inputs:
- sampler - the denoising algorithm. If you're new to this and on an SDXL-family model, the concepts doc's safe default is still DPM++ 2M Karras; on flow-matching models the picture is different, so check what your checkpoint is built on before assuming.
- scheduler - the noise schedule shape. Karras concentrates denoising effort mid-trajectory, which helps DDPM-style models and hurts many newer ones.
Outputs: sampler and scheduler, both strings. Wire them into the matching inputs of the MultiModel samplers and you've got one control point for the whole graph. That's the real use case: one place to change the sampler for every chained model at once instead of touching each one.
Install and notes
It ships with comfyui-rapidfire - no extra dependencies, no models to download, since it's pure config plumbing. Get the pack via ComfyUI Manager (search "comfyui-rapidfire") or:
cd ComfyUI/custom_nodes
git clone https://github.com/Zeks/comfyui-rapidfire
then restart ComfyUI.
One honest caveat: this node only earns its place if you're already inside the rapidfire sampler ecosystem. If you're running a stock KSampler graph, a lone sampler dropdown isn't a problem you have - the standard KSampler takes its values as widget settings, not wires. Reach for it when the graph has samplers that do take wire inputs and you want to stop editing them in several nodes at once.
Inputs (2)
| Name | Type | Default | Description |
|---|---|---|---|
| sampler | COMBO | 44 options: euler, euler_cfg_pp, euler_ancestral, euler_ancestral_cfg_pp, heun, heunpp2, +38 | |
| scheduler | COMBO | 9 options: simple, sgm_uniform, karras, exponential, ddim_uniform, beta, +3 |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| sampler | euler,euler_cfg_pp,euler_ancestral,euler_ancestral_cfg_pp,heun,heunpp2,exp_heun_2_x0,exp_heun_2_x0_sde,dpm_2,dpm_2_ancestral,lms,dpm_fast,dpm_adaptive,dpmpp_2s_ancestral,dpmpp_2s_ancestral_cfg_pp,dpmpp_sde,dpmpp_sde_gpu,dpmpp_2m,dpmpp_2m_cfg_pp,dpmpp_2m_sde,dpmpp_2m_sde_gpu,dpmpp_2m_sde_heun,dpmpp_2m_sde_heun_gpu,dpmpp_3m_sde,dpmpp_3m_sde_gpu,ddpm,lcm,ipndm,ipndm_v,deis,res_multistep,res_multistep_cfg_pp,res_multistep_ancestral,res_multistep_ancestral_cfg_pp,gradient_estimation,gradient_estimation_cfg_pp,er_sde,seeds_2,seeds_3,sa_solver,sa_solver_pece,ddim,uni_pc,uni_pc_bh2 | — |
| scheduler | * | — |