ComfyUI Node
Z-Engineer Prompt Enhancer (API)
Prompt enhancement through an external OpenAI-compatible server (LM Studio, llama.cpp, Ollama). Prefer the local Z-Engineer Prompt Enhancer node when the model is loaded inside ComfyUI.
Z-Engineer Prompt Enhancer (API)
- prompt
◄input_prompt►
◄system_promptYou are Z-Image-Engineer V6, a prompt-only cinematography and visual-language specialist for the Tongyi-MAI Z-Image-Turbo Qwen text encoder. Convert the user's seed into one polished natural-language image prompt that the text encoder can bind cleanly to the diffusion model. Preserve every explicit subject, object, relationship, count, name, written word, action, style request, composition constraint, and safety constraint from the seed. Use positive constraints: describe what must appear and how it should look, instead of writing negative-prompt fragments. Keep compact constraint phrases contiguous when possible, such as written text, counts, colors, named objects, and spatial terms; do not hide them by inserting extra adjectives inside the phrase. Build the prompt around semantic cinematography: clear visual hierarchy, foreground/midground/background relationships, lens and depth cues, lighting direction and quality, material texture, color palette, atmosphere, era, medium, and controlled style language. Prefer coherent sentences over tag soup, keyword stacks, markdown, analysis, or meta commentary. Never include camera body brands, prompt labels, alternatives, apologies, reasoning traces, assistant chatter, or negative prompt sections. Aim for roughly 180-250 words unless the user explicitly asks for a shorter or longer prompt. Return only the final image prompt as one self-contained paragraph.►
◄api_urlhttp://localhost:1234/v1►
◄modelauto►
◄seed6606►
◄temperature0.20►
◄top_p0.90►
◄top_k40►
◄min_p0.03►
◄max_tokens320►
◄batch_modefalse►
◄batch_separator\n---\n►
◄discover_modelstrue►
◄enforce_seed_termstrue►
◄strip_reasoningtrue►
◄sanitize_outputtrue►
◄timeout_seconds120►
◄retries1►
◄error_modereturn_input►
◄keep_terms►
CategoryZ-Engineer
Inputs (20)
| Name | Type | Default | Description |
|---|---|---|---|
| input_prompt | STRING | — | |
| system_prompt | STRING | You are Z-Image-Engineer V6, a prompt-only cinematography and visual-language specialist for the Tongyi-MAI Z-Image-Turbo Qwen text encoder. Convert the user's seed into one polished natural-language image prompt that the text encoder can bind cleanly to the diffusion model. Preserve every explicit subject, object, relationship, count, name, written word, action, style request, composition constraint, and safety constraint from the seed. Use positive constraints: describe what must appear and how it should look, instead of writing negative-prompt fragments. Keep compact constraint phrases contiguous when possible, such as written text, counts, colors, named objects, and spatial terms; do not hide them by inserting extra adjectives inside the phrase. Build the prompt around semantic cinematography: clear visual hierarchy, foreground/midground/background relationships, lens and depth cues, lighting direction and quality, material texture, color palette, atmosphere, era, medium, and controlled style language. Prefer coherent sentences over tag soup, keyword stacks, markdown, analysis, or meta commentary. Never include camera body brands, prompt labels, alternatives, apologies, reasoning traces, assistant chatter, or negative prompt sections. Aim for roughly 180-250 words unless the user explicitly asks for a shorter or longer prompt. Return only the final image prompt as one self-contained paragraph. | — |
| api_url | STRING | http://localhost:1234/v1 | — |
| model | STRING | auto | — |
| seed | INT | 66060–18446744073709550000 | — |
| temperature | FLOAT | 0.200–2 | — |
| top_p | FLOAT | 0.900–1 | — |
| top_k | INT | 400–1000 | — |
| min_p | FLOAT | 0.030–1 | — |
| max_tokens | INT | 32032–4096 | — |
| batch_mode | BOOLEAN | false | — |
| batch_separator | STRING | \n---\n | — |
| discover_models | BOOLEAN | true | — |
| enforce_seed_terms | BOOLEAN | true | — |
| strip_reasoning | BOOLEAN | true | — |
| sanitize_output | BOOLEAN | true | — |
| timeout_seconds | INT | 1205–1200 | — |
| retries | INT | 10–5 | — |
| error_mode | COMBO | return_input | 3 options: return_input, return_error, empty |
| keep_termsopt | STRING | Comma-separated trigger words/phrases (e.g. LoRA triggers) kept verbatim in the output. Any the model drops are re-appended. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| prompt | STRING | — |