SamplerCustomAdvanced OmniGen2Transformer2DModel.forward() missing 1 required positional argument: 'num_tokens'

#4
by ndrjvvr - opened

I'm using your workflow with your quantized model omnigen2-q5_k_m.gguf and Qwen2.5-VL-7B-instruct-Q6_K.gguf and I'm getting the forementioned error.
Should I try to resolve it or is this model not worth it anymore?
I wanted to give it a shot.

Owner

for omnigen2, you should use qwen2.5-vl-3b as text encoder instead of 7b

Oh, I thought it'll work with 7b, downloaded 3b and it's working now, thank you!

Sign up or log in to comment