[Desktop] weights failed to load: Gemma-4-26B-A4B NVFP4

[yun201720]

Before you start

• I have checked the supported models and this checkpoint is not listed there. A listed checkpoint that fails is a bug; use a Bug report instead.
• I have read the Roadmap and this model is not already planned there.
• I have read the FAQ.
• I have searched existing issues and found no duplicate.
• This is not a GGUF checkpoint. GGUF support is on the Roadmap; please do not open an issue for it.

Hugging Face link

https://huggingface.co/nvidia/Gemma-4-26B-A4B-NVFP4

Is the model architecture already supported

Yes, but this checkpoint or quantization does not load

Is the quantization already supported

Yes, but this checkpoint's weight format does not load

What happens when you load it

>>>>>>>> PASTE THE FULL SERVER LOG HERE <<<<<<<<
(copied with the "Copy server log" button in the failure dialog, or Logs -> Server status -> Copy)

WeightLoadError: OSError: 页面文件太小，无法完成操作。 (os error 1455)

Anything else

Desktop 0.2.0-beta.21; engine unknown; NVIDIA GeForce RTX 5070 Laptop GPU, 7.96 GiB, driver 591.74; Intel Core Ultra 9 285H, 31.5 GiB

[benhadad]

The error OSError: [WinError 1455] The paging file is too small for this operation to complete occurs because your Windows system's Virtual Memory (Paging File/Pagefile) is insufficient.

Is this a bug or an issue with not having Paging files setup that are large enough? @yun201720

[trcwebdesign]

You could try my fork
https://github.com/trcwebdesign/FreeToken
Windows NVFP4 Fork Enhancements
This fork adds Windows-focused support for serving large Qwen3.8 and Gemma4 NVFP4/GGUF checkpoints on consumer hardware.

Disk-tier MoE offload
--moe-disk-tier on --expert-ram-experts 128 --disk-fetch-workers 16