Ben Grasset

DiffusionFanatic1's activity

• All
• Models
• Datasets
• Spaces
• Buckets
• Papers
• Collections
• Community
• Posts
• Upvotes
• Likes
• Articles

New activity in black-forest-labs/flux-3-image-klein-comfy 5 days ago

The MOSS Image prompt enhancer run in "Caption" mode on a good LLM produces prompts that work significantly better on this than it's own prompt enhancer's "T2I" mode, in basically all cases.

👍 1

3

#41 opened 5 days ago by

DiffusionFanatic1

New activity in briaai/Fibo-2.0-dev 6 days ago

The model seems good based on the HF space but nobody is going to use it without ComfyUI support for actual local inference

1

#5 opened 6 days ago by

DiffusionFanatic1

New activity in black-forest-labs/flux-3-image-klein-comfy 10 days ago

Did you upload the wrong weights?

👀👍 12

34

#7 opened 11 days ago by

crombobularBoop

New activity in DavidAU/Qwen3.5-9B-The-Defiant-Fable-DARK-ROAST-Uncensored-Heretic-NEO-IMATRIX-MAX-MTP 2 months ago

Incoherent

1

#2 opened 2 months ago by

DiffusionFanatic1

New activity in fancyfeast/bigasp-3 2 months ago

Feedback on new Turbo Lora: VERY green everything, super heavy 80s / 90s aesthetic

4

#7 opened 2 months ago by

slightlyoutofphase

New activity in LiquidAI/LFM2.5-2.6B-GGUF 2 months ago

BF16 vs FP16?

1

#7 opened 2 months ago by

DiffusionFanatic1

New activity in LiquidAI/LFM2.5-2.6B 2 months ago

Any plans for a dense 4B?

#14 opened 2 months ago by

DiffusionFanatic1

New activity in Jackrong/Qwopus3.5-4B-Coder 4 months ago

How is this different / better than the original Qwopus3.5-4B? Nothing on the page make that clear.

#2 opened 4 months ago by

DiffusionFanatic1

New activity in google/gemma-4-E2B-it 6 months ago

How to best prompt for transcription results when sending audio input?

1

#30 opened 6 months ago by

DiffusionFanatic1

New activity in WWTCyberLab/gemma-4-E2B-it-abliterated 7 months ago

GGUF when?

🔥 1

1

#1 opened 7 months ago by

DiffusionFanatic1

New activity in ggml-org/gguf-my-repo 7 months ago

Transformers needs to be updated to support Gemma4

👍 2

11

#202 opened 7 months ago by

shakedzy

New activity in HauhauCS/Gemma-4-E2B-Uncensored-HauhauCS-Aggressive 7 months ago

I can ONLY see K_P quants both in the hardware compatibility matrix and Files And Versions?

#2 opened 7 months ago by

DiffusionFanatic1

New activity in unsloth/gemma-4-E2B-it-GGUF 7 months ago

New uploads to add llama.cpp fixes

3

#3 opened 7 months ago by

danielhanchen

New activity in unsloth/Qwen3.5-9B-GGUF 8 months ago

Enabling or disabling reasoning (default is disabled)

🔥 8

20

#2 opened 8 months ago by

danielhanchen

New activity in unsloth/Qwen3.5-4B-GGUF 8 months ago

Qwen3.5-4B runs at 50% speed of Qwen3-4B on my machine

➕ 3

12

#4 opened 8 months ago by

PurelySelfMade

New activity in circlestone-labs/Anima 8 months ago

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

👍🤯 12

8

#52 opened 8 months ago by

Schelluwu123

New activity in circlestone-labs/Anima 9 months ago

some feedback

4

#13 opened 9 months ago by

conki

is any kind of regional attention supported?

2

#8 opened 9 months 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?

2

#1 opened 9 months ago by

DiffusionFanatic1

New activity in Linum-AI/linum-v2-720p 9 months 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

#1 opened 9 months ago by

DiffusionFanatic1