13 8

Ben Grasset

DiffusionFanatic1

AI & ML interests

None yet

Recent Activity

new activity 8 days ago

circlestone-labs/Anima:plz drop the LLM Adapter. It bottlenecks Qwen 0.6B's true potential (NLU & Chinese)

liked a model 15 days ago

fal/flux-2-klein-4b-spritesheet-lora

liked a model 15 days ago

black-forest-labs/FLUX.2-klein-4B

View all activity

Organizations

None yet

New activity in circlestone-labs/Anima 8 days ago

plz drop the LLM Adapter. It bottlenecks Qwen 0.6B's true potential (NLU & Chinese)

👍 🤯 12

#52 opened 9 days ago by

Schelluwu123

New activity in circlestone-labs/Anima 23 days ago

some feedback

#13 opened 23 days ago by

conki

is any kind of regional attention supported?

#8 opened 23 days ago by

shivshankar

Maybe add a not that `lighttaew2_1.pth` is needed in the `vae_approx` folder to get high-quality TAESD previews in ComfyUI?

#1 opened 24 days ago by

DiffusionFanatic1

New activity in Linum-AI/linum-v2-720p about 1 month ago

The stated VRAM requirements on Github seem almost impossibly high for both versions of the model relative to the claimed parameter count

👍 2

#1 opened about 1 month ago by

DiffusionFanatic1

New activity in KBlueLeaf/HDM-xut-340M-anime 3 months ago

Is there any way to get a normal latent preview while running this in ComfyUI?

#2 opened 3 months ago by

DiffusionFanatic1

New activity in purplesmartai/pony-v7-base 4 months ago

FP8?

#3 opened 4 months ago by

notafraud

New activity in Qwen/Qwen-Image 4 months ago

It's so unsafe that it can generate inappropriate adult content easily

🤯 1

#21 opened 7 months ago by

liougehooa

New activity in tencent/SRPO 6 months ago

BF16 safetensors?

👍 2

#2 opened 6 months ago by

DiffusionFanatic1

New activity in fancyfeast/bigaspv2-5 7 months ago

Time to ditch SDXL?

😔 👍 7

#1 opened 7 months ago by

natalie5

New activity in Qwen/Qwen-Image 7 months ago

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

#51 opened 7 months ago by

DiffusionFanatic1

New activity in neta-art/Neta-Lumina 7 months ago

SD.Next also supports running Lumina 2.0 models

👍 4

#1 opened 7 months ago by

DiffusionFanatic1

SD.Next also supports running Lumina 2.0 models

👍 4

#1 opened 7 months ago by

DiffusionFanatic1

Ben Grasset

AI & ML interests

Recent Activity

Organizations

DiffusionFanatic1's activity

plz drop the LLM Adapter. It bottlenecks Qwen 0.6B's true potential (NLU & Chinese)

some feedback

is any kind of regional attention supported?

Maybe add a not that `lighttaew2_1.pth` is needed in the `vae_approx` folder to get high-quality TAESD previews in ComfyUI?

The stated VRAM requirements on Github seem almost impossibly high for both versions of the model relative to the claimed parameter count

Is there any way to get a normal latent preview while running this in ComfyUI?

FP8?

It's so unsafe that it can generate inappropriate adult content easily

BF16 safetensors?

Time to ditch SDXL?

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

SD.Next also supports running Lumina 2.0 models

SD.Next also supports running Lumina 2.0 models

Ben Grasset

AI & ML interests

Recent Activity

Organizations

DiffusionFanatic1's activity

plz drop the LLM Adapter. It bottlenecks Qwen 0.6B's true potential (NLU & Chinese)

some feedback

is any kind of regional attention supported?

Maybe add a not that `lighttaew2_1.pth` is needed in the `vae_approx` folder to get high-quality TAESD previews in ComfyUI?

The stated VRAM requirements on Github seem almost impossibly high for both versions of the model relative to the claimed parameter count

Is there any way to get a normal latent preview while running this in ComfyUI?

FP8?

It's so unsafe that it can generate inappropriate adult content easily

BF16 safetensors?

Time to ditch SDXL?

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

SD.Next also supports running Lumina 2.0 models

SD.Next also supports running Lumina 2.0 models