Billing SKU mismatch: gemini-2.5-flash-lite input charged correctly but output charged as "gemini 2.5 flash" (not lite)

Gemini API
1

1. Rate Limits Page (Correct)

The Google AI Studio Rate Limits page shows I’m using gemini-2.5-flash-lite:

  • Model: gemini-2.5-flash-lite

  • Category: Text-out models

  • RPM: 1 / 4K

  • TPM: 34.08K / 4M

2. Billing Report (Mismatch)

SKU Usage Cost
gemini 2.5 flash lite short input text 92,797 tokens $0.29
gemini 2.5 flash short output text non-thinking 123,298 tokens $1.55

The output cost of $1.55 for 123,298 tokens equals ~$12.57/1M tokens, which is much higher than:

  • Expected Flash Lite output: $0.40/1M

  • Even Flash output: $2.50/1M

My Code

from google import genai

client = genai.Client(api_key=api_key)
model_name = "gemini-2.5-flash-lite"  # Also tried "models/gemini-2.5-flash-lite"

response = client.models.generate_content(
    model=model_name,
    contents=contents
)

Environment

  • SDK: google-genai (latest version)

  • Model: gemini-2.5-flash-lite

  • API: Google AI Studio (not Vertex AI)

Questions

  1. Why is the output being billed under a different SKU (gemini 2.5 flash) than the input (gemini 2.5 flash lite)?

  2. Is there any additional configuration needed to ensure output tokens are also billed at the Flash Lite rate?

  3. Is this a known billing system issue?

Thank you for your help!