Spark-X2.6
A conversation that thinks, on your own CPU
XHToken's Spark-X2.6-1.7B, a text chat model, compiled to WebAssembly with fusor-ai. Ask it anything, let it reason in a thinking block first or answer straight away, and read the reply as it is written. Once loaded, nothing you type leaves the page.
- First token
- from Send to the reply's start
- Speed
- tokens written per second
- Tokens
- in the reply
- Context
- Network
- requests since the model loaded