Google Imagen Models (Deprecated)

All Google Imagen model IDs are deprecated and no longer listed in data/models.json, so do not use the older imagen-* examples or the v1beta :predict image route for new AvalAI integrations. This page remains only to help developers migrate old bookmarks and code snippets.

Deprecated Models

Deprecated model	Replacement
imagen-4.0-ultra-generate-001	gemini-3-pro-image (Nano Banana Pro)
imagen-4.0-generate-001	gemini-3.1-flash-image (Nano Banana 2)
imagen-4.0-fast-generate-001	gemini-3.1-flash-image (Nano Banana 2)
imagen-3.0-generate-002	gemini-3.1-flash-lite-image (Nano Banana 2 Lite)
imagen-3.0-generate-001	gemini-3.1-flash-lite-image (Nano Banana 2 Lite)
imagen-3.0-fast-generate-001	gemini-3.1-flash-lite-image (Nano Banana 2 Lite)

What to Use Instead

• Nano Banana series: Use gemini-3.1-flash-image, gemini-3-pro-image, or gemini-3.1-flash-lite-image for Google image generation and editing through v1/chat/completions or the native v1beta :generateContent endpoint. (gemini-2.5-flash-image shuts down on October 2, 2026, so prefer the Gemini 3.x image models for new integrations.)
• GPT Image: Use gpt-image-2 for high-quality generation, editing, text rendering, compositing, and mask-based edits through v1/images/generations and v1/images/edits.
• Other providers: FLUX (flux.2-pro, flux.1-kontext-pro), Qwen Image, and Seedream remain available for direct image generation workloads.

Migration Example

Old Imagen native-API call:

# Deprecated - do not use
curl -X POST \
  "https://api.avalai.ir/v1beta/models/imagen-4.0-fast-generate-001:predict" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"instances": [{"prompt": "A serene Japanese garden"}], "parameters": {"sampleCount": 1}}'

New Nano Banana call through the OpenAI-compatible endpoint:

Bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "gemini-3.1-flash-image",
    "messages": [
      {"role": "user", "content": "Generate a serene Japanese garden with a koi pond"}
    ],
    "modalities": ["image", "text"]
  }'

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AVALAI_API_KEY"],
    base_url="https://api.avalai.ir/v1",
)

response = client.chat.completions.create(
    model="gemini-3.1-flash-image",
    messages=[
        {"role": "user", "content": "Generate a serene Japanese garden with a koi pond"}
    ],
    modalities=["image", "text"],
)

images = response.choices[0].message.images
if images:
    print("Image URL:", images[0]["image_url"]["url"])

JavaScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AVALAI_API_KEY,
  baseURL: "https://api.avalai.ir/v1",
});

const response = await client.chat.completions.create({
  model: "gemini-3.1-flash-image",
  messages: [
    { role: "user", content: "Generate a serene Japanese garden with a koi pond" },
  ],
  modalities: ["image", "text"],
});

const images = response.choices[0].message.images;
if (images?.length) {
  console.log("Image URL:", images[0].image_url.url);
}

For the full prompting playbook, aspect-ratio and image-size controls, and editing examples, see the Nano Banana series guide and the model deprecations and migration guide.

For model availability, always check data/models.json through the Models API or the model browser.