Nice! Can you add a message saying it doesn't work on browsers that don't support it? You can also support using SmolLM2-360M-Instruct as fallback for those browsers. I made https://github.com/mkornreich/browser_llm to get that logic working if you want to use it.
Running large models in the browser doesn't seem like a good idea—even after a long download, the model still isn't ready. Running Ollama locally is probably a better choice.
Nice! Can you add a message saying it doesn't work on browsers that don't support it? You can also support using SmolLM2-360M-Instruct as fallback for those browsers. I made https://github.com/mkornreich/browser_llm to get that logic working if you want to use it.
Running large models in the browser doesn't seem like a good idea—even after a long download, the model still isn't ready. Running Ollama locally is probably a better choice.
You successfully attracted me with the Cheap name. However it looks like slow to me.
[dead]
[flagged]
[dead]