ComfyUI Node
LingBot TI2V Sampler (First Frame Fixed)
A ComfyUI node in LingBot Video with 10 inputs and 1 output.
LingBot TI2V Sampler (First Frame Fixed)
- model
- conditioning
- latents
◄width640►
◄height352►
◄num_frames73►
◄steps40►
◄cfg3.0►
◄shift3.00►
◄seed42►
◄cfg_executionsequential_sage►
CategoryLingBot Video
Inputs (10)
| Name | Type | Default | Description |
|---|---|---|---|
| model | LINGBOT_MODEL | — | |
| conditioning | LINGBOT_CONDITIONING | — | |
| width | INT | 64016–4096 | — |
| height | INT | 35216–4096 | — |
| num_frames | INT | 731–1001 | — |
| steps | INT | 401–1000 | — |
| cfg | FLOAT | 3.01–30 | — |
| shift | FLOAT | 3.000.01–20 | — |
| seed | INT | 420–18446744073709550000 | — |
| cfg_executionopt | COMBO | sequential_sage | Sequential Sage is the conservative baseline. Batched native computes exact masked B=2 CFG in one pass. Sage-varlen is experimental. All batched modes auto-fallback above 16,384 video tokens or on CUDA OOM. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| latents | LINGBOT_LATENTS | — |