Candle Phi 1.5

Rust/WASM Demo

The Phi-1.5 model achieves state-of-the-art performance with only 1.3 billion parameters, compared to models with up to 10 billion. You can try the quantized version of the model here. Additional prompt examples are available in the technical report .

You can also try Puffin-Phi V2 quantized version model, a fine-tuned version of Phi-1.5 on the Puffin dataset

Note: When first run, the app will download and cache the model, which could take a few minutes. The models are ~800MB or ~1.51GB in size.

Generation: