Google 旗舰, 长上下文 + Thinking 文本输入、文本输出(含思考)和缓存命中按上下文分档。音频输入按每百万 Token 收费,原价 $1.25。
Model ID: gemini-2.5-pro · Type: chat · Provider: Google
Endpoints: /v1/chat/completions · /v1/messages · /v1beta/models/{model}:generateContent · /v1beta/models/{model}:streamGenerateContent · /v1/models/{model}:generateContent · /v1/models/{model}:streamGenerateContent · /v1/responses
| Input (per 1M tokens) | $0.9375 USD |
| Output (per 1M tokens) | $7.5 USD |
| Cache read (per 1M tokens) | $0.09375 USD |
| ≤ 200000 tokens | $0.9375 / 1M in | $7.5 / 1M out |
| > 200000 | $1.875 / 1M in | $11.25 / 1M out |
import os
from openai import OpenAI
client = OpenAI(api_key=os.environ["ROUTER_API_KEY"], base_url="https://api.router.ai/v1")
resp = client.chat.completions.create(
model="gemini-2.5-pro",
messages=[{"role": "user", "content": "Hello"}],
)
print(resp.choices[0].message.content)Gemini 2.5 Pro (`gemini-2.5-pro`) is billed per usage at $0.9375/1M in · $7.5/1M out, in USD. Current pricing is always listed at https://370.ai/models/gemini-2.5-pro.
Send a request to https://api.router.ai/v1/chat/completions with the header `Authorization: Bearer <your API key>` and `"model": "gemini-2.5-pro"`. The API is OpenAI-compatible, so any OpenAI SDK works by changing base_url to https://api.router.ai/v1 — no other code change.
Gemini 2.5 Pro can be called on: /v1/chat/completions; /v1/messages; /v1beta/models/{model}:generateContent; /v1beta/models/{model}:streamGenerateContent; /v1/models/{model}:generateContent; /v1/models/{model}:streamGenerateContent; /v1/responses.
Gemini 2.5 Pro accepts up to 1,048,576 input tokens and can return up to 65,535 output tokens. Requests exceeding the input limit are rejected before reaching the model.
Gemini 2.5 Pro supports: vision, function_calling, prompt_caching, audio_input, thinking, long_context, cache.
Gemini 2.5 Pro is a chat model from Google, available through the 370.AI gateway with the same API key as every other model.
Call it through the 370.AI OpenAI-compatible endpoint (API base: https://api.router.ai/v1). AI agents can browse the catalog and call any text model through MCP (https://mcp.router.ai/mcp); for image, video and other modalities use the API above.