ComfyUI Node: QwenVL (Advanced)
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.
Category
QwenVL-Utils
Inputs
model_name
- Qwen3.5-0.8B
- Qwen3.5-2B
- Qwen3.5-4B
- Qwen3.5-9B
- Qwen3.5-9B-FP8
- Qwen3.5-27B
- Qwen3.5-27B-FP8
- Qwen3.5-35B-A3B
- Qwen3.5-35B-A3B-FP8
- Qwen3-VL-2B-Instruct
- Qwen3-VL-2B-Thinking
- Qwen3-VL-2B-Instruct-FP8
- Qwen3-VL-2B-Thinking-FP8
- Qwen3-VL-4B-Instruct
- Qwen3-VL-4B-Thinking
- Qwen3-VL-4B-Instruct-FP8
- Qwen3-VL-4B-Thinking-FP8
- Qwen3-VL-8B-Instruct
- Qwen3-VL-8B-Thinking
- Qwen3-VL-8B-Instruct-FP8
- Qwen3-VL-8B-Thinking-FP8
- Qwen3-VL-30B-A3B-Instruct
- Qwen3-VL-30B-A3B-Thinking
- Qwen3-VL-30B-A3B-Instruct-FP8
- Qwen3-VL-30B-A3B-Thinking-FP8
- Qwen3-VL-32B-Instruct
- Qwen3-VL-32B-Thinking
- Qwen3-VL-32B-Instruct-FP8
- Qwen3-VL-32B-Thinking-FP8
- Qwen2.5-VL-3B-Instruct
- Qwen2.5-VL-3B-Instruct-AWQ
- Qwen2.5-VL-7B-Instruct
- Qwen2.5-VL-7B-Instruct-AWQ
- Qwen2.5-VL-32B-Instruct
- Qwen2.5-VL-32B-Instruct-AWQ
- Qwen2-VL-2B-Instruct
- Qwen2-VL-2B-Instruct-AWQ
- Qwen2-VL-7B-Instruct
- Qwen2-VL-7B-Instruct-AWQ
- [GGUF] Qwen2.5-VL-32B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen2.5-VL-32B-Instruct-Q8_0.gguf
- [GGUF] Qwen2.5-VL-3B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen2.5-VL-3B-Instruct-Q8_0.gguf
- [GGUF] Qwen2.5-VL-7B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen2.5-VL-7B-Instruct-Q8_0.gguf
- [GGUF] Qwen3.5-0.8B-BF16.gguf
- [GGUF] Qwen3.5-0.8B-Q4_K_M.gguf
- [GGUF] Qwen3.5-0.8B-Q8_0.gguf
- [GGUF] Qwen3.5-27B-BF16.gguf
- [GGUF] Qwen3.5-27B-Q4_K_M.gguf
- [GGUF] Qwen3.5-27B-Q8_0.gguf
- [GGUF] Qwen3.5-2B-BF16.gguf
- [GGUF] Qwen3.5-2B-Q4_K_M.gguf
- [GGUF] Qwen3.5-2B-Q8_0.gguf
- [GGUF] Qwen3.5-35B-A3B-BF16.gguf
- [GGUF] Qwen3.5-35B-A3B-Q4_K_M.gguf
- [GGUF] Qwen3.5-35B-A3B-Q8_0.gguf
- [GGUF] Qwen3.5-4B-BF16.gguf
- [GGUF] Qwen3.5-4B-Q4_K_M.gguf
- [GGUF] Qwen3.5-4B-Q8_0.gguf
- [GGUF] Qwen3.5-9B-BF16.gguf
- [GGUF] Qwen3.5-9B-Q4_K_M.gguf
- [GGUF] Qwen3.5-9B-Q8_0.gguf
- [GGUF] Qwen3VL-2B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen3VL-2B-Instruct-Q8_0.gguf
- [GGUF] Qwen3VL-2B-Thinking-Q4_K_M.gguf
- [GGUF] Qwen3VL-2B-Thinking-Q8_0.gguf
- [GGUF] Qwen3VL-30B-A3B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen3VL-30B-A3B-Instruct-Q8_0.gguf
- [GGUF] Qwen3VL-30B-A3B-Thinking-Q4_K_M.gguf
- [GGUF] Qwen3VL-30B-A3B-Thinking-Q8_0.gguf
- [GGUF] Qwen3VL-32B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen3VL-32B-Instruct-Q8_0.gguf
- [GGUF] Qwen3VL-32B-Thinking-Q4_K_M.gguf
- [GGUF] Qwen3VL-32B-Thinking-Q8_0.gguf
- [GGUF] Qwen3VL-4B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen3VL-4B-Instruct-Q8_0.gguf
- [GGUF] Qwen3VL-4B-Thinking-Q4_K_M.gguf
- [GGUF] Qwen3VL-4B-Thinking-Q8_0.gguf
- [GGUF] Qwen3VL-8B-Instruct-Q4_K_M.gguf
- [GGUF] Qwen3VL-8B-Instruct-Q8_0.gguf
- [GGUF] Qwen3VL-8B-Thinking-Q4_K_M.gguf
- [GGUF] Qwen3VL-8B-Thinking-Q8_0.gguf
enable_thinking BOOLEAN
quantization
- 4-bit (VRAM-friendly)
- 8-bit (Balanced)
- None (FP16)
attention_mode
- auto
- flash_attention_2
- sdpa_flash
- sage_attention
- sdpa_math
- eager
- sdpa
use_torch_compile BOOLEAN
device
- auto
- cpu
- mps
preset_prompt
- ❌ None
- 🖼️ Tags
- 🖼️ Simple Description
- 🖼️ Detailed Description
- 🖼️ Ultra Detailed Description
- 🎬 Cinematic Description
- 🖼️ Detailed Analysis
- 📹 Video Summary
- 📖 Short Story
- 🪄 Prompt Refine & Expand
custom_prompt STRING
max_tokens INT
temperature FLOAT
top_p FLOAT
num_beams INT
repetition_penalty FLOAT
frame_count INT
keep_model_loaded BOOLEAN
seed INT
image IMAGE
video IMAGE
source_path PATH
min_pixels INT
max_pixels INT
ctx INT
n_batch INT
gpu_layers INT
image_max_tokens INT
top_k INT
min_p FLOAT
top_k_sampling INT
pool_size INT
Outputs
STRING
Extension: ComfyUI-QwenVL-Utils
Comprehensive QwenVL integration for ComfyUI with HuggingFace and GGUF model support
Authored by AkihaTatsu
Looking for a different node?
More nodes in ComfyUI-QwenVL-Utils
Run ComfyUI workflows without the setup
No installs, no CUDA version roulette, no GPU sitting idle on your bill. Bring a workflow and run it in the browser.