Gemini 2.5 Series Stable Models Released: New Model Names Available
Date: 2025-06-28
Summary
Google has officially released stable versions of the Gemini 2.5 series models. The new stable model names gemini-2.5-flash and gemini-2.5-pro are now available, replacing the previous preview versions. Additionally, we've added the new gemma-3n-e2b-it model to the Gemma series and the experimental gemini-2.5-flash-lite-preview-06-17 model.
Details
This update brings significant improvements to the Gemini 2.5 series with the release of stable production-ready models. Users should migrate from preview versions to the new stable model names for better reliability and performance.
Google Gemini 2.5 Stable Models
gemini-2.5-flash: The stable version of Google's fast and efficient Gemini 2.5 model, optimized for speed while maintaining high-quality outputs. This replaces the preview versions and offers improved stability for production use. Documentation
gemini-2.5-pro: The stable version of Google's most capable Gemini 2.5 model, offering advanced reasoning and comprehensive language understanding. This is the production-ready version of the previously available preview models. Documentation
Migration Guide for Preview Models
If you're currently using preview model names, please update your code to use the new stable versions:
Gemini 2.5 Pro Migration
- From:
gemini-2.5-pro-preview-06-05→ To:gemini-2.5-pro - From:
gemini-2.5-pro-preview-05-06→ To:gemini-2.5-pro
Gemini 2.5 Flash Migration
- From:
gemini-2.5-flash-preview-04-17→ To:gemini-2.5-flash - From:
gemini-2.5-flash-preview-05-20→ To:gemini-2.5-flash
New Gemma Model
- gemma-3n-e2b-it: A new addition to the Gemma series, featuring enhanced capabilities and optimized performance for instruction-following tasks. Documentation
New Experimental Model
- gemini-2.5-flash-lite-preview-06-17: An experimental lightweight version of Gemini 2.5 Flash, designed for applications requiring even faster response times with reduced computational requirements. Documentation
Usage Examples
Using the New Stable Gemini 2.5 Models
from openai import OpenAI
client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")
# Using the stable Gemini 2.5 Flash model
completion = client.chat.completions.create(
model="gemini-2.5-flash",
messages=[
{
"role": "user",
"content": "Explain the benefits of using stable model versions over preview versions.",
}
],
)
print(completion.choices[0].message.content)
# Using the stable Gemini 2.5 Pro model
completion = client.chat.completions.create(
model="gemini-2.5-pro",
messages=[
{
"role": "user",
"content": "Analyze the impact of AI model stability on production applications.",
}
],
)
print(completion.choices[0].message.content)import { OpenAI } from "openai";
const client = new OpenAI({
apiKey: process.env.AVALAI_API_KEY,
baseURL: "https://api.avalai.ir/v1",
});
// Using the stable Gemini 2.5 Flash model
const flashCompletion = await client.chat.completions.create({
model: "gemini-2.5-flash",
messages: [
{
role: "user",
content:
"Explain the benefits of using stable model versions over preview versions.",
},
],
});
console.log(flashCompletion.choices[0].message.content);
// Using the stable Gemini 2.5 Pro model
const proCompletion = await client.chat.completions.create({
model: "gemini-2.5-pro",
messages: [
{
role: "user",
content:
"Analyze the impact of AI model stability on production applications.",
},
],
});
console.log(proCompletion.choices[0].message.content);Using the New Gemma Model
from openai import OpenAI
client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")
completion = client.chat.completions.create(
model="gemma-3n-e2b-it",
messages=[
{
"role": "user",
"content": "Write a detailed explanation of machine learning concepts for beginners.",
}
],
)
print(completion.choices[0].message.content)import { OpenAI } from "openai";
const client = new OpenAI({
apiKey: process.env.AVALAI_API_KEY,
baseURL: "https://api.avalai.ir/v1",
});
const completion = await client.chat.completions.create({
model: "gemma-3n-e2b-it",
messages: [
{
role: "user",
content:
"Write a detailed explanation of machine learning concepts for beginners.",
},
],
});
console.log(completion.choices[0].message.content);