TensorStack

/

Flux_schnell-f16-onnx

like 2

Follow

TensorStack AI 33

Hello, could you please tell me how you exported the ONNX? When I use torch.onnx.export(pipe.transformer, (latents, prompt_embeds, pooled_prompt_embeds, timestep, latent_image_ids, text_ids, guidance, joint_attention_kwargs, False), "transformer.onnx", opset_version=16), the screen always prints a large number of tensors, which seems to be due to the model being too large. Have you encountered this issue?