katuni4ka commited on
Commit
1347630
·
verified ·
1 Parent(s): 1b3357d

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +34 -2
README.md CHANGED
@@ -18,7 +18,8 @@ The provided OpenVINO™ IR model is compatible with:
18
  * OpenVINO version 2024.1.0 and higher
19
  * Optimum Intel 1.16.0 and higher
20
 
21
- ## Running Model Inference
 
22
 
23
  1. Install packages required for using [Optimum Intel](https://huggingface.co/docs/optimum/intel/index) integration with the OpenVINO backend:
24
 
@@ -36,7 +37,7 @@ model_id = "OpenVINO/starcoder2-15b-fp16-ov"
36
  tokenizer = AutoTokenizer.from_pretrained(model_id)
37
  model = OVModelForCausalLM.from_pretrained(model_id)
38
 
39
- inputs = tokenizer("def print_hello_world()", return_tensors="pt")
40
 
41
  outputs = model.generate(**inputs, max_length=200)
42
  text = tokenizer.batch_decode(outputs)[0]
@@ -45,6 +46,37 @@ print(text)
45
 
46
  For more examples and possible optimizations, refer to the [OpenVINO Large Language Model Inference Guide](https://docs.openvino.ai/2024/learn-openvino/llm_inference_guide.html).
47
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
48
  ## Legal information
49
 
50
  The original model is distributed under [bigcode-openrail-m](https://www.bigcode-project.org/docs/pages/bigcode-openrail/) license. More details can be found in [bigcode/starcoder2-15b](https://huggingface.co/bigcode/starcoder2-15b).
 
18
  * OpenVINO version 2024.1.0 and higher
19
  * Optimum Intel 1.16.0 and higher
20
 
21
+ ## Running Model Inference with [Optimum Intel](https://huggingface.co/docs/optimum/intel/index)
22
+
23
 
24
  1. Install packages required for using [Optimum Intel](https://huggingface.co/docs/optimum/intel/index) integration with the OpenVINO backend:
25
 
 
37
  tokenizer = AutoTokenizer.from_pretrained(model_id)
38
  model = OVModelForCausalLM.from_pretrained(model_id)
39
 
40
+ inputs = tokenizer("def print_hello_world():", return_tensors="pt")
41
 
42
  outputs = model.generate(**inputs, max_length=200)
43
  text = tokenizer.batch_decode(outputs)[0]
 
46
 
47
  For more examples and possible optimizations, refer to the [OpenVINO Large Language Model Inference Guide](https://docs.openvino.ai/2024/learn-openvino/llm_inference_guide.html).
48
 
49
+ ## Running Model Inference with [OpenVINO GenAI](https://github.com/openvinotoolkit/openvino.genai)
50
+
51
+ 1. Install packages required for using OpenVINO GenAI.
52
+ ```
53
+ pip install openvino-genai huggingface_hub
54
+ ```
55
+
56
+ 2. Download model from HuggingFace Hub
57
+
58
+ ```
59
+ import huggingface_hub as hf_hub
60
+
61
+ model_id = "OpenVINO/starcoder2-15b-fp16-ov"
62
+ model_path = "starcoder2-15b-fp16-ov"
63
+
64
+ hf_hub.snapshot_download(model_id, local_dir=model_path)
65
+
66
+ ```
67
+
68
+ 3. Run model inference:
69
+
70
+ ```
71
+ import openvino_genai as ov_genai
72
+
73
+ device = "CPU"
74
+ pipe = ov_genai.LLMPipeline(model_path, device)
75
+ print(pipe.generate("def print_hello_world():"))
76
+ ```
77
+
78
+ More GenAI usage examples can be found in OpenVINO GenAI library [docs](https://github.com/openvinotoolkit/openvino.genai/blob/master/src/README.md) and [samples](https://github.com/openvinotoolkit/openvino.genai?tab=readme-ov-file#openvino-genai-samples)
79
+
80
  ## Legal information
81
 
82
  The original model is distributed under [bigcode-openrail-m](https://www.bigcode-project.org/docs/pages/bigcode-openrail/) license. More details can be found in [bigcode/starcoder2-15b](https://huggingface.co/bigcode/starcoder2-15b).