microsoft
/

Phi-4-multimodal-instruct

Automatic Speech Recognition

text-generation

speech-summarization

speech-translation

visual-question-answering

phi-4-multimodal

Model card Files Files and versions Community

Resources

View closed (43)

Error when loading audio processor saved with processor.save_pretrained(path_to_dir), and num_logits_to_keep wrong init value

#73 opened about 9 hours ago by

When yhou support an arabic audio?

#72 opened 13 days ago by

cannot import name 'SlidingWindowCache' from 'transformers.cache_utils'

#71 opened 14 days ago by

Convert to HF format

#70 opened 20 days ago by

Inference with an external LM?

#69 opened 21 days ago by

Phi4 output gibberish

#68 opened 25 days ago by

Error when loading model quantized with BitsAndBytesConfig for inference

#67 opened 27 days ago by

Speech to Vision?

#66 opened 28 days ago by

DDP error when multi-gpu finetuning with speech encoder parameters unfrozen

#63 opened 29 days ago by

Finetuning Vision part for phi 4 mm

#62 opened about 1 month ago by

fix loading processor after save_pretrained with transformers 4.49+

#60 opened about 1 month ago by

Quantized Fine-Tunning

#59 opened about 1 month ago by

transformers==4.48.2

#58 opened about 1 month ago by

vLLM Deployment Problems

#57 opened about 1 month ago by

Suggested command fails to start with vLLM 0.8.1

#51 opened about 1 month ago by

Decoding strategy of the Phi4 Multimodal

#50 opened about 2 months ago by

how to inference with vllm?

#48 opened about 2 months ago by

Fine-tuning on a new language

#47 opened about 2 months ago by

[Bug] Fixed NoneType Slicing Error in `modeling_phi4mm.py`: “bad operand type for unary -”

#46 opened about 2 months ago by

Fix image embedding logic to be mps-compatible

#45 opened about 2 months ago by

Padding of labels bug?

#44 opened about 2 months ago by

Experience with Phi-4-Multimodal vs. Whisper-1 for Speech-to-Text

#39 opened about 2 months ago by

Does the model support beam search for ASR?

#31 opened 2 months ago by

`tokenizer.model` is missing

#30 opened 2 months ago by

Getting Bounding Boxes for Vision

#29 opened 2 months ago by

how to input a video?

#16 opened 2 months ago by

Error during inference with image and text.

#12 opened 2 months ago by

llama.cpp support when?

#7 opened 2 months ago by

How to use it with LM Studio?

#3 opened 2 months ago by

thanks , how to fine tune?

#1 opened 2 months ago by