jadehardouin commited on
Commit
e0e93c4
1 Parent(s): f5562e5

Add application file

Browse files
Files changed (1) hide show
  1. app.py +43 -0
app.py ADDED
@@ -0,0 +1,43 @@
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
1
+ import gradio as gr
2
+
3
+ def calculate_tco(model_choice, vm_rental_choice):
4
+ VM_cost_per_hour=3.6730
5
+ tokens_per_request = 64
6
+
7
+ if model_choice == "Llama-2-7B":
8
+ tokens_per_second=694.38
9
+
10
+ elif model_choice == "Llama-2-13B":
11
+ tokens_per_second=1000
12
+
13
+ elif model_choice == "Llama-2-70B":
14
+ tokens_per_second=10000
15
+
16
+ if vm_rental_option == "1 month" or "3 months" or "6 months":
17
+ reduction = 0
18
+
19
+ elif vm_rental_option == "1 year":
20
+ reduction = 0.34
21
+
22
+ elif vm_rental_option == "3 years":
23
+ reduction = 0.62
24
+
25
+ homemade_cost_per_request = VM_cost_per_hour * (1 - reduction) * tokens_per_request / (tokens_per_second * 3600) #* 0.8 * 0.5
26
+ saas_cost_per_request = 0.018 * tokens_per_request
27
+
28
+ output = f"Cost/request with a plan for {vm_rental_choice} using:\n"
29
+ output += f"- SaaS solution from OpenAI: ${saas_cost_per_request:.5f}\n"
30
+ output += f"- Home-made solution with the model {model_choice}: ${homemade_cost_per_request:.5f}"
31
+ return output
32
+
33
+ models = ["Llama-2-7B", "Llama-2-13B", "Llama-2-70B"]
34
+ vm_rental_option = ["1 month", "3 months", "6 months", "1 year", "3 years"]
35
+
36
+ iface = gr.Interface(
37
+ fn=calculate_tco,
38
+ inputs=[gr.inputs.Dropdown(models, label="Select AI Model"),
39
+ gr.inputs.Dropdown(vm_rental_option, label="Select VM Rental Duration")],
40
+ outputs=gr.outputs.Textbox()
41
+ )
42
+
43
+ iface.launch()