Niels G's picture

24

Niels G

NielsGx

AI & ML interests

None yet

Organizations

None yet

New activity in QuantStack/Wan2.2-Animate-14B-GGUF 3 months ago

Size difference compared to Kijai version

#3 opened 3 months ago by

New activity in Kijai/WanVideo_comfy_GGUF 3 months ago

Different size compared to QuantStack version

#11 opened 3 months ago by

New activity in lightx2v/Qwen-Image-Lightning 4 months ago

Cannot use torch.compile with that LoRA

#5 opened 5 months ago by

New activity in Qwen/Qwen-Image-Edit 4 months ago

HW requirements

#6 opened 4 months ago by

Releasing FP8 & F16 Models

#9 opened 4 months ago by

New activity in lightx2v/Qwen-Image-Lightning 5 months ago

Is 8step model higher quality than the 4step one ?

#4 opened 5 months ago by

New activity in Qwen/Qwen-Image 5 months ago

What range of resolutions is the model actually trained in, in terms of megapixels?

#51 opened 5 months ago by

DiffusionFanatic1

New activity in nunchaku-tech/nunchaku-t5 5 months ago

umt-t5 ?

#2 opened 5 months ago by

New activity in Kijai/WanVideo_comfy_fp8_scaled 5 months ago

What is the difference between this and ComfyOrg's quantitative model?

#5 opened 5 months ago by

How to use fp8 mat-mul?

#7 opened 5 months ago by

New activity in Kijai/WanVideo_comfy 5 months ago

Issue with Wan2_1-T2V-14B_fp8 model. Images look unfinished even after 50 steps.

#55 opened 5 months ago by

Wan-Lighting : 4steps per model or 4steps total?

#59 opened 5 months ago by

New activity in unsloth/Qwen2.5-VL-7B-Instruct-GGUF 5 months ago

UD vs non-UD ?

#1 opened 5 months ago by

New activity in Kijai/WanVideo_comfy 5 months ago

FP16 Lora on FP8 scaled model

#52 opened 5 months ago by

New activity in lightx2v/Wan2.1-I2V-14B-720P-StepDistill-CfgDistill-Lightx2v 5 months ago

Empty repo?

#1 opened 5 months ago by

New activity in jt-zhang/SageAttention3 5 months ago

Any improvement on Ada Lovelace (RTX 4xxx) ?

#1 opened 5 months ago by

New activity in vrgamedevgirl84/Wan14BT2VFusioniX 6 months ago

Request to replace Causvid with lightx2v Self Forcing CFG Distill (better than Causvid)

#18 opened 6 months ago by

New activity in LifuWang/DistillT5 9 months ago

Use with HiDream-I1 diffusion model ?

#2 opened 9 months ago by

New activity in azaneko/HiDream-I1-Full-nf4 9 months ago

Use original text encoders alongside NF4 transformer model (mix and match)

#3 opened 9 months ago by

New activity in zhang0jhon/flux_wavelet 9 months ago

FP8 or GGUF version to run on consumer GPU

#1 opened 9 months ago by