0.0.18

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.2.3) for comfyui (0.3.13)
• pig model support added; new workflow inside
• others; please refer to comfyui

0.0.17

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.2.1) for comfyui (0.3.13)
• embedded engine gguf-connector upgraded to the latest version (1.1.5)
• embedded python site-packages: pytorch (2.6) cuda (12.6) upgraded
• better error messages for common errors; etc.
• others; please refer to comfyui

0.0.16

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.2.0) for comfyui (0.3.12)
• pixart model supported for both safetensors and gguf
• tensor cutter (beta), gguf convertor (alpha) and gguf convertor (reverse) implemented
• embedded engine gguf-connector upgraded to the latest version (1.1.4)
• others; please refer to comfyui

0.0.15

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.1.7) for comfyui (0.3.12)
• new feature: gguf convertor (reverse); able to convert gguf back to safetensors model
• tensor cutter (beta) and gguf convertor (alpha) implemented; quantize safetensors file and/or convert it to gguf
• embedded engine gguf-connector upgraded to the latest version (1.1.3)
• others; please refer to comfyui

0.0.14

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.1.6) for comfyui (0.3.12)
• tensor cutter and gguf convertor implemented
• cosmos, hyvid, and mochi for gguf clip loader supported
• embedded engine gguf-connector upgraded to the latest version (1.1.2)
• others; please refer to comfyui

0.0.13

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.1.4) for comfyui (0.3.12)
• tensor cutter (beta) and gguf convertor (alpha) implemented; ease to quantize and make your own model
• cosmos, mochi, and hyvid for gguf clip loader support added; restructure the data workflow; bug fixed; etc.
• others; please refer to comfyui

0.0.12

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.1.2) for comfyui (0.3.12)
• hyvid for gguf dual clip loader supported; can use llava-llama3 gguf with scaled clip-l as text encoders right away (workflow inside)
• cosmos and mochi for gguf loaders added; working in progress
• tensor cutter (beta) and gguf convertor (alpha) implemented
• others; please refer to comfyui

0.0.11

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.1.1) for comfyui (0.3.12)
• cosmos and mochi support for gguf loaders added
• tensor cutter (beta) and gguf convertor (alpha) implemented
• checkpoints are always loaded safely on pytorch 2.4 and above; oom issue fixed; etc.
• others; please refer to comfyui

0.0.10

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.0.8) for comfyui (0.3.12)
• tensor cutter (beta) and convertor (alpha) implemented; DIY tools: create/make your own quantized model(s) easy; waiting others' feed and knee down for help are no longer needed; just do it yourself
• nvidia cosmos 7B and 14B: text2video and image/video2video diffusion models supported (with workflow inside)
• ckpt/pt/etc. files are always loaded safely on pytorch 2.4 and above now; bug fixed; etc.
• others; please refer to comfyui

0.0.9

ComfyUI_GGUF_windows_portable

key highlights:

• gguf-node (1.0.7) for comfyui (0.3.11)
• tensor cutter (beta) and convertor (alpha) implemented; DIY: create/make your own quantized model(s) possible
• new sampler: res_multistep added; Nvidia Cosmos 7B & 14B: text to video and image to video model(s) supported
• some cases of ancestral samplers not being deterministic fixed, etc.
• others; please refer to comfyui