Sipp
WebGPU Chat
Model
Choose model
Text
No model loaded
New chat
Settings
Backend
--
Decode
--
Start a conversation
Load a local model
Explain WebGPU
Pink lemonade
Compare inference
Remove
Add image
Send
Stop
Local inference
Choose a model
Close
Curated
Custom
URL
Local file
GGUF model URL
GGUF model file
No file selected
Custom imports support text models. Choose a curated bundle for vision.
Generation
Settings
Close
Max tokens
Maximum response length
Temperature
Higher values increase variation
0.70
Top-p
Nucleus sampling threshold
0.90