Manual Tracking | Paygent Documentation

Manual tracking gives you full control when you already know the exact token counts. Use manual tracking when:

✓ You’re calling AI APIs directly and have access to usage data
✓ You want maximum control over what’s reported
✓ You’re tracking non-OpenAI models or custom services
✓ You have your own token counting logic

Example: OpenAI Integration

Python

Node.js

Go

manual_openai.py

1 from paygent_sdk import Client, UsageData
2 import os
3 from openai import OpenAI
4 
5 paygent = Client.new_client(os.getenv("PAYGENT_API_KEY"))
6 openai_client = OpenAI(api_key=os.getenv("OPENAI_API_KEY"))
7 
8 # Make OpenAI API call
9 response = openai_client.chat.completions.create(
10     model="gpt-4o",
11     messages=[{"role": "user", "content": "Explain quantum physics"}]
12 )
13 
14 # Extract usage data
15 usage = response.usage
16 cached_tokens = 0
17 if hasattr(usage, 'prompt_tokens_details'):
18     cached_tokens = usage.prompt_tokens_details.cached_tokens or 0
19 
20 # Track with Paygent
21 usage_data = UsageData(
22     service_provider="OpenAI",
23     model=response.model,
24     prompt_tokens=usage.prompt_tokens,
25     completion_tokens=usage.completion_tokens,
26     cached_tokens=cached_tokens,
27     total_tokens=usage.total_tokens
28 )
29 
30 paygent.send_usage("qa-agent", "customer-456", "question-answered", usage_data)

Automate Onboarding. To simplify your workflow, call client.create_or_get_customer() before tracking usage. This ensures the customer exists in Paygent without needing manual setup in the dashboard.

Example: Gemini Integration

Python

Node.js

Go

manual_gemini.py

1 from paygent_sdk import Client, UsageData
2 import os
3 import google.generativeai as genai
4 
5 paygent = Client.new_client(os.getenv("PAYGENT_API_KEY"))
6 genai.configure(api_key=os.getenv("GOOGLE_API_KEY"))
7 
8 # Make Gemini API call
9 model = genai.GenerativeModel('gemini-2.0-flash-exp')
10 response = model.generate_content('Write a poem about AI')
11 
12 # Extract usage data
13 metadata = response.usage_metadata
14 
15 # Track with Paygent
16 usage_data = UsageData(
17     service_provider="Google DeepMind",
18     model="gemini-2.0-flash-exp",
19     prompt_tokens=metadata.prompt_token_count,
20     completion_tokens=metadata.candidates_token_count,
21     total_tokens=metadata.total_token_count
22 )
23 
24 paygent.send_usage("content-gen", "customer-789", "poem-created", usage_data)