gemini-2.5-flash — multi-supplier realtime bidding · OpenAI + Anthropic protocol parity · get a key
What does the gemini-2.5-flash API cost and how do I call it? gemini-2.5-flash is input $0.0072 and output $0.06 per 1M tokens at the market floor across 17 competing suppliers (data date 2026-11-20; machine-readable: https://321ai.cc/api/public/model-prices). Set base_url to https://321ai.cc/v1 for the OpenAI protocol or ANTHROPIC_BASE_URL to https://321ai.cc for Claude Code. Failures are never billed; platform fee is 0.
| Item | Market floor price (USD / 1M tokens) |
|---|---|
| Input | $0.0072 |
| Output | $0.06 |
| Cache read | $0.00072 |
Floor across suppliers; per-supplier quotes on the marketplace. Machine-readable: https://321ai.cc/models/gemini-2.5-flash.md · all models: /en/models/.
from openai import OpenAI
client = OpenAI(api_key="YOUR_KEY", base_url="https://321ai.cc/v1")
resp = client.chat.completions.create(
model="gemini-2.5-flash",
messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)
Anthropic protocol (Claude Code): export ANTHROPIC_BASE_URL=https://321ai.cc + ANTHROPIC_AUTH_TOKEN=YOUR_KEY. Guide: Anthropic Messages API via 321AI.
gemini-2.5-flash is billed per token: input around $0.0072 / 1M tokens, output around $0.06 / 1M tokens, cache reads around $0.00072 / 1M tokens. This page shows the market floor; per-supplier quotes are on the marketplace page.
Yes — fully OpenAI API compatible. Set base_url to https://321ai.cc/v1 with your platform API key; existing OpenAI SDKs and common clients work without code changes.
17 suppliers currently quote gemini-2.5-flash on 321AI with transparent price comparison; merchants publish their own upstream self-check records on the marketplace page.
Yes. New accounts get free trial credit — call gemini-2.5-flash from the console before you top up. Sign up: https://321ai.cc/ui/new/#/register