ComfyUI Node Runs on cloud
Generate Edit Instruction
Turn a short edit request into an explicit instruction an image editing model follows fully, written by a language model that reads the picture: what changes, implied effects included, and what stays. With a judge's notes wired in, it corrects the last attempt.
Generate Edit Instruction
- clip
- image
- instruction
◄instruction►
◄temperature0.00►
◄seed0►
◄max_length256►
◄feedback—►
◄last_instruction—►
CategoryWAS Suite/Text
Inputs (8)
| Name | Type | Default | Description |
|---|---|---|---|
| clip | CLIP | qwen3.5_9b_int8_convrot.safetensors from Load CLIP with type krea2, or another Qwen-VL model. | |
| image | IMAGE | The picture to edit. The first of a batch is read. | |
| instruction | STRING | The edit as written, as 'Make it a snowy winter night'. | |
| temperature | FLOAT | 0.000–2 | 0 = the likeliest words, the same instruction every time; 0.7 = varied wording per seed. |
| seed | INT | 00–18446744073709550000 | The seed sampling draws from. Ignored at temperature 0. |
| max_length | INT | 25632–2048 | The most tokens the instruction may run to: 256 covers about 180 words. |
| feedbackopt | STRING | Judge Images' notes on the last attempt, as 'the sky is blown out and shows no snow'. Unwired or empty = a first attempt. | |
| last_instructionopt | STRING | The instruction the last attempt was made from, as 'Add falling snow and a dark blue sky'. |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| instruction | STRING | The explicit instruction, for the edit model's prompt and Judge Images. |