microsoft
/

Phi-3-mini-4k-instruct-onnx-web

Text Generation

Transformers.js

Model card Files Files and versions Community

Resources

View closed (1)

Update Transformers.js config to use fp16 kv cache for q4f16 model

#3 opened about 16 hours ago by

Shape incompatibility when ```past_key_values``` are made persistent for context retention

#2 opened 8 months ago by