cellm

Model

Tokenizer:

Backend:

Waiting for model file...

VIEW LOGS

Max tokens:

Tokens

0

Sessions

0

Free Blocks

0

Send a message to start the conversation.
Load a VLM model + enable WebGPU for image chat.

Image (JPEG/PNG):

Max tokens:

Upload an image and click Describe to generate a description...

Vision encoder: -ms | Decode: -ms | Total: -ms