microsoft
/

Phi-4-mini-instruct

Text Generation

text-generation-inference

Inference Endpoints

Model card Files Files and versions Community

Resources

View closed (7)

pipeline for text generation seems to be giving garbage outputs

#12 opened about 2 hours ago by

[Bug]:Phi-4-Mini giving garbage outputs with torch 2.5.1 and vllm==0.7.3 with multiple parallel requests on Long context prompts #14058

#11 opened about 9 hours ago by

Suggested tokenizer changes similar to Phi-4

#8 opened 3 days ago by

Thanks!

#3 opened 4 days ago by

Different number of attention heads, makes rotary_ndims vs rope scaling factors wrong?

#1 opened 5 days ago by