Hugging Face's logo

Xenova
/
Phi-3-mini-4k-instruct

Text Generation
Transformers.js
ONNX
phi3
ONNX
DML
ONNXRuntime
nlp
conversational
custom_code
Model card Files Community
5
Xenova HF staff commited on
Commit
ed2beb4
•
1 Parent(s): 9f76e6d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +7 -9
README.md CHANGED
@@ -1,14 +1,13 @@
1
  ---
2
  license: mit
3
  pipeline_tag: text-generation
 
4
  tags:
5
- - ONNX
6
- - DML
7
- - ONNXRuntime
8
- - phi3
9
- - nlp
10
- - conversational
11
- - custom_code
12
  ---
13
 
14
  # Phi-3 Mini-128K-Instruct ONNX model for onnxruntime-web
@@ -17,5 +16,4 @@ This is the same models as the [official phi3 onnx model](https://huggingface.co
17
  1. the model is fp16 with int4 block quantization for weights
18
  2. the 'logits' output is fp32
19
  3. the model uses MHA instead of GQA
20
- 4. onnx and external data file need to stay below 2GB to be cacheable in chromium
21
-
 
1
  ---
2
  license: mit
3
  pipeline_tag: text-generation
4
+ library_name: transformers.js
5
  tags:
6
+ - ONNX
7
+ - DML
8
+ - ONNXRuntime
9
+ - nlp
10
+ - conversational
 
 
11
  ---
12
 
13
  # Phi-3 Mini-128K-Instruct ONNX model for onnxruntime-web
 
16
  1. the model is fp16 with int4 block quantization for weights
17
  2. the 'logits' output is fp32
18
  3. the model uses MHA instead of GQA
19
+ 4. onnx and external data file need to stay below 2GB to be cacheable in chromium