Cheap Gemini API access

Gemini model prices

Workload: a chatbot with 10,000 requests a month (1,500 input + 500 output tokens each). Effective prices use the minimum 2× credit; top-ups of $50, $100 and $200 get 2.2×, 2.35× and 2.5×.

How to cut Gemini API cost further

• Start with Gemini 3.8 Flash and only move up when quality demands it — Gemini 3.1 Pro costs 3.0× more per blended token.
• Keep system prompts short and reuse them so cached input is billed at the lower cache rate.
• Cap output length with max_tokens; output tokens cost several times more than input.
• Top up $200 at once for the 2.5× tier instead of many small top-ups.

Cheaper alternatives to Gemini on the same key

Switch by changing only the model name.

Call Gemini through Somnus

from openai import OpenAI
import os

client = OpenAI(
    api_key=os.environ["SOMNUS_API_KEY"],
    base_url="https://gateway-production-c837.up.railway.app/v1",
)

resp = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[
      {"role": "user", "content": "Hello Gemini"}
    ],
)
print(resp.choices[0].message.content)

Pick the model name from your dashboard.

Questions

Related