Search results

TypeError in _get_perc_reserved_mem_max: '<=' not supported between instances of 'str' and 'int'

Status: Open.

#27 In deepbeepmeep/mmgp;

· antoninononooono opened on Feb 28, 2026

Trellis 2 launched few days ago and the optimization on it is weird

Status: Open.

#24 In deepbeepmeep/mmgp;

· thruCD opened on Dec 20, 2025

[IMPROVEMENTS] Add some impovements

Status: Open.

#23 In deepbeepmeep/mmgp;

· SlimRG opened on Dec 20, 2025

help with optimizing this new app Sam3video from facebook

Status: Open.

#22 In deepbeepmeep/mmgp;

· manat0912 opened on Dec 15, 2025

offload.py

Status: Open.

#21 In deepbeepmeep/mmgp;

· tmpHero opened on Nov 27, 2025

Qwen-Image Support?

Status: Open.

#20 In deepbeepmeep/mmgp;

· canytam-krystal opened on Oct 31, 2025

does it work with intel gpu?

Status: Open.

#17 In deepbeepmeep/mmgp;

· bangboomtata opened on Oct 24, 2025

'TinyGemmWeightQBitsTensor' object has no attribute '_scale'

Status: Open.

#16 In deepbeepmeep/mmgp;

· wikeeyang opened on Oct 19, 2025

Direct3d s2

Status: Open.

#14 In deepbeepmeep/mmgp;

· thruCD opened on Jul 14, 2025

can't convert cuda:0 device type tensor to numpy. Use Tensor.cpu() to copy the tensor to host memory first. When attempting to call a diffusers sdxl pipeline after doing inference with the mmgp library active.

Status: Open.

#12 In deepbeepmeep/mmgp;

· Meatfucker opened on Jun 20, 2025

The confusion about the usage of MMGP

Status: Open.

#9 In deepbeepmeep/mmgp;

· lixiangcog opened on Apr 20, 2025

mmgp vs accelerate

Status: Open.

#7 In deepbeepmeep/mmgp;

· okaris opened on Apr 7, 2025