Tortoise17
Tortoise17
@wenhuach21 yes, that exception has been given. Due to that reason the process started for the pickup of the checkpoints. The below is now complete log. `2025-09-29 18:36:27 INFO autoround.py...
seqlen should be set at which stage? ``` ar = AutoRound( model=model_name_or_path, scheme="W4A16", iters=50, lr=5e-3, seqlen=512, ) ``` like this didnt work
@wenhuach21 @n1ck-guo in my tests, RTN MODE for exporting autoround format works so far. But, while exporting, there is error or log warning `2025-09-30 10:24:09 INFO base.py L751: thinker.visual.blocks.0.mlp.linear_fc1 will...
@wenhuach21 @allerou4 [Qwen3-Omni Quantized](https://huggingface.co/cpatonn/Qwen3-Omni-30B-A3B-Instruct-AWQ-4bit) Here is the link for the Qwen3-Omni. But, I have found out that with auto-round pipelines are more stable than other mode of techniques. I am...
@allerou4 memory wise that is not stable as well. vllm and transformers are also culprit.
@wenhuach21 The current code and model needs 360 GB vRAM which is much higher than normal resources demand.
@comfyui-wiki This happened to me only for this template. Maybe I am messing up something. If you can hint the mistake or ignorance.
@comfyui-wiki fixed it finally. Thank you. Just a request. If you can share the template which could have the animation node with options batch_images so that I can animate the...
@comfyui-wiki can you share this template json?