ComfyUI Node
NIMbus Vision (PMS)
A ComfyUI node in PromptModels/NVIDIA (NIMbus) with 9 inputs and 1 output.
NIMbus Vision (PMS)
- image
- text
◄promptDescribe this image in detail.►
◄modelnvidia/nemotron-nano-12b-v2-vl►
◄system_prompt►
◄custom_model►
◄enable_thinkingfalse►
◄temperature0.70►
◄max_tokens1024►
◄api_key►
CategoryPromptModels/NVIDIA (NIMbus)
Inputs (9)
| Name | Type | Default | Description |
|---|---|---|---|
| image | IMAGE | Imagen para análisis multimodal (VL models). | |
| prompt | STRING | Describe this image in detail. | — |
| model | COMBO | nvidia/nemotron-nano-12b-v2-vl | 3 options: nvidia/nemotron-nano-12b-v2-vl, nvidia/llama-3.1-nemotron-nano-vl-8b-v1, nvidia/vila |
| system_promptopt | STRING | Instrucción de sistema. Vacío = sin system. | |
| custom_modelopt | STRING | Sobreescribe el combo si no está vacío. | |
| enable_thinkingopt | BOOLEAN | false | Activa razonamiento extendido (nvext.thinking). Solo modelos que lo soporten. |
| temperatureopt | FLOAT | 0.700–2 | — |
| max_tokensopt | INT | 102464–8192 | — |
| api_keyopt | STRING | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |