Models / Google

Model vendor

Google

Google DeepMind makes Imagen, Veo and the Gemini family of multimodal long-context models with native image and video support.

52Models routed
+265%Average vs list price
4Channels serving Google

Google models on MirAPI

ModelTypeContextInputOutputCache r / wOff listChannels
Gemini 4 ArgonnewText1.049M$2.00$10.00— / —soon—
Nano Banana 2.1newImage66K$1.50$7.50— / —soon—
Gemini 3.8 Flash Lite TTSnewSpeech33K$0.50—— / —soon—
Gemini 3.8 Flash TTSnewSpeech33K$0.50—— / —soon—
Chirp 3Transcription—$266.67—— / —n/a—
Veo 3.1 FastVideo——$0.080per second of output · 720p— / —n/a—
Veo 3.1 LiteVideo——$0.030per second of output · 720p— / —n/a—
Lyria 3 Clip PreviewAudio1.049M——— / —n/a—
Lyria 3 Pro PreviewAudio1.049M——— / —n/a—
Veo 3.1Video——$0.200per second of output · 720p— / —n/a—
Gemini 2.5 Pro Preview 06-05Text1.049M$1.25$10.00— / —n/a—
Gemma 3 12BText131K$0.05$0.15— / —n/a—
Gemma 3 27BText131K$0.08$0.35— / —n/a—
Gemma 3 4BText131K$0.05$0.10— / —n/a—
Gemma 2 27BText8K$0.65$0.65— / —n/a—
Browse the full catalog
Migration

Gemini through the OpenAI-compatible API

Call Gemini, Imagen and Veo through the same OpenAI-compatible endpoint as every other model. You don’t need a Google Cloud project or service accounts, only the model ID from our catalog:

Endpoints
/v1/chat/completionsChat & tool calls
/v1/embeddingsText embeddings
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.mirapi.ai/v1",   # changed
    api_key=os.environ["MIRAPI_API_KEY"],   # changed
)

resp = client.chat.completions.create(
    model="gemma-4-31b-it",
    messages=[{"role": "user", "content": "Hello"}],
)

One key covers Google and 10 more vendors.

1M free tokens to start. No card required.