Nodes/VLM_nodes/Molmo Vision-Language Model
ComfyUI Node Runs on cloud

Molmo Vision-Language Model

A ComfyUI node in VLM Nodes/Molmo with 9 inputs and 1 output.

By gokayfem·Created 3 years ago·Updated 4 days ago· 578
Molmo Vision-Language Model
  • image
  • STRING
promptDescribe this image in detail.
model_name
memory_mode4-bit Quantized (15GB+ Required)
max_new_tokens200
temperature0.7
top_p0.9
top_k50
use_autocasttrue
CategoryVLM Nodes/Molmo

Inputs (9)

NameTypeDefaultDescription
imageIMAGEInput image to be analyzed by Molmo
promptSTRINGDescribe this image in detail.Instructions for the model. Be specific about what aspects of the image you want analyzed.
model_nameCOMBO⚠️ WARNING: These are very large models requiring significant RAM/VRAM. Start with MolmoE-1B.
memory_modeCOMBO4-bit Quantized (15GB+ Required)Controls RAM/VRAM usage. Use most aggressive option that works on your system.
max_new_tokensINT2001–2048Maximum tokens to generate. Higher values need more VRAM. Start small (200) and increase if needed.
temperatureFLOAT0.70.1–2Controls randomness. Lower (0.1-0.5) = more focused, higher (0.8-2.0) = more creative.
top_pFLOAT0.90.1–1Nucleus sampling. Lower = more focused on likely tokens, higher = more diverse vocabulary.
top_kINT501–100Limits token choices to top K most likely. Lower = more focused, higher = more variety.
use_autocastBOOLEANtrueEnables mixed precision. Keeps quality while reducing VRAM usage. Recommended ON.

Outputs (1)

NameTypeDescription
STRINGSTRING