125 2

Silver PRO

silveroxides

silveroxides

AI & ML interests

None yet

Recent Activity

new activity 42 minutes ago

Comfy-Org/Qwen-Image_ComfyUI:Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

new activity 42 minutes ago

Comfy-Org/Qwen-Image_ComfyUI:Upload mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

new activity 43 minutes ago

Comfy-Org/Qwen-Image_ComfyUI:Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

View all activity

Organizations

New activity in Comfy-Org/Qwen-Image_ComfyUI 42 minutes ago

Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

#15 opened about 1 hour ago by

silveroxides

Upload mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

#14 opened 20 days ago by

silveroxides

New activity in Comfy-Org/Qwen-Image_ComfyUI 43 minutes ago

Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

#16 opened 43 minutes ago by

silveroxides

New activity in silveroxides/FLUX.2-dev-fp8_scaled 13 days ago

fp8 scaled discussion

#2 opened 21 days ago by

Muawiz

New activity in stduhpf/Chroma1-Radiance-gguf 15 days ago

Doesn't Work

#1 opened 22 days ago by

qpqpqpqpqpqp

New activity in silveroxides/Qwen-Image-fp8-scaled-quants 15 days ago

Exception during processing !!! Activation scale shape mismatch: scale.shape=torch.Size([96, 24]), expected (12288, 24)

#1 opened 16 days ago by

MonsterMMORPG

New activity in Comfy-Org/Qwen-Image-Edit_ComfyUI 15 days ago

Upload MixedPrecisionOps compatible fp8mixed version of Qwen Image Edit 2511

#14 opened 15 days ago by

silveroxides

New activity in silveroxides/Qwen3-4B-ComfyQuants-SingleFiles 19 days ago

Size??

#1 opened 26 days ago by

clover-supply

New activity in silveroxides/Chroma1-HD-fp8-scaled 19 days ago

Benchmarks

#2 opened 20 days ago by

jtreminio

New activity in Comfy-Org/Qwen-Image-Layered_ComfyUI 20 days ago

Upload mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

#2 opened 20 days ago by

silveroxides

New activity in silveroxides/Z-Image-De-Turbo-fp8_scaled 24 days ago

Update README.md (Fix)

#1 opened 25 days ago by

qpqpqpqpqpqp

New activity in Comfy-Org/Qwen-Image-Edit_ComfyUI 24 days ago

Add MixedPrecisionOps compatible fp8mixed qwen_image_edit_2509 with sensitive layers kept in high precision and using full_precision_matrix_mult

#11 opened 24 days ago by

silveroxides

New activity in kakkkarotto/whatever 2 months ago

Upload dasiwaWAN22I2V14B_radiantcrushLow-diffonly.safetensors with huggingface_hub

#2 opened 2 months ago by

silveroxides

Upload dasiwaWAN22I2V14B_radiantcrushHigh-diffonly.safetensors with huggingface_hub

#1 opened 2 months ago by

silveroxides

New activity in silveroxides/Chroma1-HD-fp8-scaled 2 months ago

Can we get an FP8 based off FP not BF?

#1 opened 3 months ago by

TheNecr0mancer

New activity in lodestones/Chroma1-HD 3 months ago

Update model card with paths to workflow and sample and overview

#19 opened 3 months ago by

silveroxides

Upload workflow, sample image and overview image for ComfyUI

#18 opened 3 months ago by

silveroxides

New activity in lodestones/Chroma 3 months ago

Update paths for the workflow, sample image and overview

#121 opened 3 months ago by

silveroxides

New updated workflow with overview and sample

#120 opened 3 months ago by

silveroxides

Delete before uploading updated one

#119 opened 3 months ago by

silveroxides

Silver PRO

AI & ML interests

Recent Activity

Organizations

silveroxides's activity

Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

Upload mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

Upload updated mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

fp8 scaled discussion

Doesn't Work

Exception during processing !!! Activation scale shape mismatch: scale.shape=torch.Size([96, 24]), expected (12288, 24)

Upload MixedPrecisionOps compatible fp8mixed version of Qwen Image Edit 2511

Size??

Benchmarks

Upload mixed precision fp8 quantized model with comfy_quant layer configs and sensitive layers kept in high precision

Update README.md (Fix)

Add MixedPrecisionOps compatible fp8mixed qwen_image_edit_2509 with sensitive layers kept in high precision and using full_precision_matrix_mult

Upload dasiwaWAN22I2V14B_radiantcrushLow-diffonly.safetensors with huggingface_hub

Upload dasiwaWAN22I2V14B_radiantcrushHigh-diffonly.safetensors with huggingface_hub

Can we get an FP8 based off FP not BF?

Update model card with paths to workflow and sample and overview

Upload workflow, sample image and overview image for ComfyUI

Update paths for the workflow, sample image and overview

New updated workflow with overview and sample

Delete before uploading updated one

🎉 Free Image Generator Now Available!