ComfyUI Node
H3 Segment Prompter
A ComfyUI node in H3 Planner with 22 inputs and 2 outputs.
H3 Segment Prompter
- project
- cast
- timeline
- report
◄h3_prompt►
◄total_seconds60.0►
◄segment_seconds10.0►
◄shots_per_segment1►
◄formatauto►
◄audio_roleperformed on camera►
◄providerOllama (Local)►
◄ollama_urlhttp://127.0.0.1:11434►
◄ollama_modelqwen3-vl:8b►
◄temperature0.25►
◄reuse_existingtrue►
◄seed0►
◄idea►
◄style_prefix_override►
◄api_key►
◄api_model►
◄max_output_tokens3072►
◄num_ctx8192►
◄keep_alive10m►
◄request_timeout600►
CategoryH3 Planner
Inputs (22)
| Name | Type | Default | Description |
|---|---|---|---|
| project | H3_PROJECT | — | |
| h3_prompt | STRING | the h3_prompt output of your H3 prompt creator, run at the FULL video duration. Leave empty to plan from `idea` instead. | |
| total_seconds | FLOAT | 60.01–600 | the whole video's length — the same number you gave the prompt creator |
| segment_seconds | FLOAT | 10.01–15 | rough length per segment; only used when there is no treatment to cut on |
| shots_per_segment | INT | 11–12 | 1 = one segment per shot in the treatment, and one prompt written for it — the most detail per clip. Raise it to pack several shots into one longer clip while they still fit under the segment cap. |
| format | COMBO | auto | what kind of video this is. Steers emphasis per segment; auto adds nothing. |
| audio_role | COMBO | performed on camera | who makes the sound in the connected audio. Set 'performed on camera' when the subject raps, sings or speaks it and their mouth must match the words - without that H3 plays the track over someone sitting silently. Ignored when no audio asset is in the cast. |
| provider | COMBO | Ollama (Local) | 1 options: Ollama (Local) |
| ollama_url | STRING | http://127.0.0.1:11434 | — |
| ollama_model | STRING | qwen3-vl:8b | — |
| temperature | FLOAT | 0.250–1.2 | — |
| reuse_existing | BOOLEAN | true | skip segments already written from the same inputs |
| seed | INT | 00–4294967295 | change to rewrite every segment |
| castopt | H3_CAST | — | |
| ideaopt | STRING | extra direction applied to every segment; the whole plan when h3_prompt is empty | |
| style_prefix_overrideopt | STRING | blank = lift it off [Shot 1] of the treatment | |
| api_keyopt | STRING | prefer the environment variable — a key typed here is saved into the workflow | |
| api_modelopt | STRING | — | |
| max_output_tokensopt | INT | 3072256–32768 | — |
| num_ctxopt | INT | 81922048–131072 | — |
| keep_aliveopt | STRING | 10m | — |
| request_timeoutopt | INT | 60030–3600 | — |
Outputs (2)
| Name | Type | Description |
|---|---|---|
| timeline | H3_TIMELINE | — |
| report | STRING | — |