ComfyUI Node
🧪 QwenVL Multi-Image (Advanced)
A ComfyUI node in 🧪AILab/QwenVL with 16 inputs and 1 output.
🧪 QwenVL Multi-Image (Advanced)
- images
- images_batch_2
- images_batch_3
- text
◄model_nameQwen/Qwen3-VL-4B-Instruct►
◄system_promptYou are a helpful assistant.►
◄user_promptDescribe these images in detail.►
◄quantization8-bit (Balanced)►
◄max_tokens1024►
◄temperature0.7►
◄top_p0.90►
◄top_k50►
◄repetition_penalty1.1►
◄num_beams1►
◄keep_model_loadedtrue►
◄seed1►
◄deviceauto►
Category🧪AILab/QwenVL
Inputs (16)
| Name | Type | Default | Description |
|---|---|---|---|
| images | IMAGE | — | |
| model_name | COMBO | Qwen/Qwen3-VL-4B-Instruct | 13 options: Qwen/Qwen3-VL-4B-Instruct, Qwen/Qwen3-VL-8B-Instruct, Qwen/Qwen3-VL-32B-Instruct, Qwen/Qwen3-VL-8B-Thinking, Qwen/Qwen3-VL-32B-Thinking, Qwen/Qwen3-VL-8B-Instruct-FP8, +7 |
| system_prompt | STRING | You are a helpful assistant. | — |
| user_prompt | STRING | Describe these images in detail. | — |
| quantization | COMBO | 8-bit (Balanced) | 3 options: None (FP16), 8-bit (Balanced), 4-bit (VRAM-friendly) |
| max_tokens | INT | 102464–4096 | — |
| temperature | FLOAT | 0.70.1–2 | — |
| top_p | FLOAT | 0.900–1 | — |
| top_k | INT | 501–100 | — |
| repetition_penalty | FLOAT | 1.11–2 | — |
| num_beams | INT | 11–10 | — |
| keep_model_loaded | BOOLEAN | true | — |
| seed | INT | 11–4294967295 | — |
| device | COMBO | auto | 3 options: auto, cuda, cpu |
| images_batch_2opt | IMAGE | — | |
| images_batch_3opt | IMAGE | — |
Outputs (1)
| Name | Type | Description |
|---|---|---|
| text | STRING | — |