Ask AI

New Flagship Gemini 3.5 Flash Added and Gemini 3.1 Flash-Lite Stable Released

Date: 2026-05-19 / (1405-02-29)

Summary

We are pleased to announce the addition of Google's new flagship multimodal reasoning model, gemini-3.5-flash, along with the stable release of gemini-3.1-flash-lite. Both models are now available through the OpenAI-compatible v1/chat/completions endpoint and the native Gemini v1beta/models endpoint.

Details

Google (Gemini)

Gemini 3.5 Flash

gemini-3.5-flash is Google's next-generation flagship Flash model, built on the Gemini 3 Flash reasoning foundation with configurable thinking levels to balance quality, cost, and latency. Published in May 2026, it delivers strong improvements in coding, agentic tool use, expert tasks, multimodal understanding, and long-context performance compared with Gemini 3 Flash.

Feature	Details
Context window	1,048,576 input tokens, 65,536 output tokens
Knowledge cutoff	January 2025
Input pricing	$1.50 / 1M tokens
Cached input pricing	$0.25 / 1M tokens
Output pricing	$9.00 / 1M tokens
Audio input pricing	$1.00 / 1M tokens
Audio cached input	$0.50 / 1M tokens
Audio output pricing	$1.00 / 1M tokens
Supported inputs	Text, Image, Video, Audio, PDF
Supported outputs	Text
Supported endpoints	`v1/chat/completions`, `v1beta/`
Thinking levels	low, medium, high

Key Features:

Flagship Flash Intelligence: Built on the Gemini 3 Flash reasoning foundation with measurable benchmark improvements
Configurable Thinking: Supports thinkingLevel (low / medium / high) to control the quality/cost/latency trade-off
Multimodal Native: Accepts text, images, audio, video, and PDF inputs in a single request
Long Context: 1M input token context window suitable for large documents and codebases
Tool Use: Native function calling, parallel function calling, structured outputs, prompt caching, and Google Search grounding
Coding & Agents: Strong results on Terminal-bench 2.1 (76.2%), SWE-Bench Pro (55.1%), MCP Atlas (83.6%), and Toolathlon (56.5%)
Multimodal Reasoning: CharXiv Reasoning 84.2%, MMMU-Pro 83.6%, OSWorld-Verified 78.4%
Reasoning: Humanity's Last Exam 40.2%, ARC-AGI-2 72.1%, MRCR v2 1M pointwise 26.6%

Best Use Cases:

High-throughput agentic workflows that need strong tool-use and coding ability at a Flash price point
Multimodal pipelines combining vision, audio, and document understanding
Long-context analysis over codebases, research papers, and structured documents
Cost-aware deployments where thinking levels can be tuned per request

Gemini 3.1 Flash-Lite (Stable Release)

gemini-3.1-flash-lite is the stable release of gemini-3.1-flash-lite-preview, previously announced in our March 10, 2026 update. The stable alias delivers the same capabilities and pricing as the preview model, providing a reliable, cost-efficient option for high-volume multimodal workloads.

Feature	Details
Context window	1,048,576 input tokens, 65,536 output tokens
Knowledge cutoff	January 2025
Input pricing	$0.25 / 1M tokens
Cached input pricing	$0.025 / 1M tokens
Output pricing	$1.50 / 1M tokens
Audio input pricing	$0.50 / 1M tokens
Audio cached input	$0.05 / 1M tokens
Audio output pricing	$1.50 / 1M tokens
Supported inputs	Text, Image, Video, Audio, PDF
Supported outputs	Text
Supported endpoints	`v1/chat/completions`, `v1beta/`

Key Features:

Production-Ready: Stable alias suitable for long-term integrations
Most Cost-Efficient Gemini 3 Model: Same low pricing as the preview release
Multimodal: Text, image, video, audio, and PDF input support
Thinking Support: Configurable thinking levels for step-by-step reasoning
Function Calling, Structured Outputs, Search Grounding, URL Context

Pricing Summary

Model	Input ($/1M tokens)	Cached Input ($/1M tokens)	Output ($/1M tokens)	Audio Input	Audio Cached	Audio Output
`gemini-3.5-flash`	$1.50	$0.25	$9.00	$1.00	$0.50	$1.00
`gemini-3.1-flash-lite`	$0.25	$0.025	$1.50	$0.50	$0.05	$1.50

API Request/Response Examples

Gemini 3.5 Flash Example (OpenAI-Compatible)

bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "gemini-3.5-flash",
    "messages": [
      {
        "role": "user",
        "content": "Outline a robust strategy to migrate a large monolithic Python service to event-driven microservices."
      }
    ]
  }'

python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.chat.completions.create(
    model="gemini-3.5-flash",
    messages=[
        {
            "role": "user",
            "content": "Outline a robust strategy to migrate a large monolithic Python service to event-driven microservices.",
        }
    ],
)

print(response.choices[0].message.content)

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-3.5-flash",
  messages: [
    {
      role: "user",
      content:
        "Outline a robust strategy to migrate a large monolithic Python service to event-driven microservices.",
    },
  ],
});

console.log(response.choices[0].message.content);

Response:

json

{
  "id": "chatcmpl-abc789",
  "created": 1747641600,
  "model": "gemini-3.5-flash",
  "object": "chat.completion",
  "system_fingerprint": null,
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "A robust migration strategy starts with the strangler-fig pattern: identify bounded contexts, expose them behind a stable API, then incrementally extract each context into an independent service while keeping the monolith authoritative until cut-over. Introduce an event backbone (e.g., Kafka or Pub/Sub) early, define outbox-based publishing for transactional integrity, and require contract tests for every new service. Plan observability, schema evolution, and rollback paths before touching production traffic.",
        "role": "assistant",
        "thinking_blocks": [],
        "annotations": []
      }
    }
  ],
  "usage": {
    "completion_tokens": 92,
    "prompt_tokens": 23,
    "total_tokens": 115,
    "completion_tokens_details": null,
    "prompt_tokens_details": {
      "audio_tokens": null,
      "cached_tokens": null,
      "text_tokens": 23,
      "image_tokens": null
    }
  },
  "estimated_cost": {
    "unit": "0.0008625000",
    "irt": 98.84,
    "exchange_rate": 114600
  }
}

Gemini 3.5 Flash with Thinking Configuration

bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "gemini-3.5-flash",
    "messages": [
      {
        "role": "user",
        "content": "A train leaves Station A at 8:00 AM traveling at 60 km/h. Another train leaves Station B at 9:00 AM traveling at 80 km/h toward Station A. The stations are 280 km apart. When and where do they meet?"
      }
    ],
    "extra_body": {
      "generationConfig": {
        "thinkingConfig": {
          "thinkingLevel": "high"
        }
      }
    }
  }'

python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.chat.completions.create(
    model="gemini-3.5-flash",
    messages=[
        {
            "role": "user",
            "content": "A train leaves Station A at 8:00 AM traveling at 60 km/h. Another train leaves Station B at 9:00 AM traveling at 80 km/h toward Station A. The stations are 280 km apart. When and where do they meet?",
        }
    ],
    extra_body={"generationConfig": {"thinkingConfig": {"thinkingLevel": "high"}}},
)

print(response.choices[0].message.content)

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-3.5-flash",
  messages: [
    {
      role: "user",
      content:
        "A train leaves Station A at 8:00 AM traveling at 60 km/h. Another train leaves Station B at 9:00 AM traveling at 80 km/h toward Station A. The stations are 280 km apart. When and where do they meet?",
    },
  ],
  // @ts-ignore - extra_body is supported by AvalAI for Gemini-specific options
  extra_body: {
    generationConfig: { thinkingConfig: { thinkingLevel: "high" } },
  },
});

console.log(response.choices[0].message.content);

Gemini 3.5 Flash via Native v1beta Endpoint

bash

curl https://api.avalai.ir/v1beta/models/gemini-3.5-flash:generateContent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          {"text": "Summarize the main advantages of event-driven architectures in three bullet points."}
        ]
      }
    ],
    "generationConfig": {
      "thinkingConfig": {"thinkingLevel": "medium"}
    }
  }'

python

from google import genai
from google.genai import types

client = genai.Client(
    api_key="your-avalai-api-key",
    http_options={"base_url": "https://api.avalai.ir"},
)

response = client.models.generate_content(
    model="gemini-3.5-flash",
    contents="Summarize the main advantages of event-driven architectures in three bullet points.",
    config=types.GenerateContentConfig(
        thinking_config=types.ThinkingConfig(thinking_level="medium"),
    ),
)

print(response.text)

javascript

import { GoogleGenAI } from "@google/genai";

const ai = new GoogleGenAI({
  apiKey: process.env.AVALAI_API_KEY,
  httpOptions: { baseUrl: "https://api.avalai.ir" },
});

const response = await ai.models.generateContent({
  model: "gemini-3.5-flash",
  contents:
    "Summarize the main advantages of event-driven architectures in three bullet points.",
  config: {
    thinkingConfig: { thinkingLevel: "medium" },
  },
});

console.log(response.text);

Gemini 3.1 Flash-Lite Stable Example

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-lite",
    "messages": [
      {
        "role": "user",
        "content": "Classify the following support ticket into one of: billing, technical, account, other. Ticket: I cannot reset my password and the email link has expired."
      }
    ]
  }'

python

from openai import OpenAI

client = OpenAI(api_key="your-avalai-api-key", base_url="https://api.avalai.ir/v1")

response = client.chat.completions.create(
    model="gemini-3.1-flash-lite",
    messages=[
        {
            "role": "user",
            "content": "Classify the following support ticket into one of: billing, technical, account, other. Ticket: I cannot reset my password and the email link has expired.",
        }
    ],
)

print(response.choices[0].message.content)

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-lite",
  messages: [
    {
      role: "user",
      content:
        "Classify the following support ticket into one of: billing, technical, account, other. Ticket: I cannot reset my password and the email link has expired.",
    },
  ],
});

console.log(response.choices[0].message.content);

Response:

json

{
  "id": "chatcmpl-def456",
  "created": 1747641900,
  "model": "gemini-3.1-flash-lite",
  "object": "chat.completion",
  "system_fingerprint": null,
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "account",
        "role": "assistant",
        "thinking_blocks": [],
        "annotations": []
      }
    }
  ],
  "usage": {
    "completion_tokens": 2,
    "prompt_tokens": 41,
    "total_tokens": 43,
    "completion_tokens_details": null,
    "prompt_tokens_details": {
      "audio_tokens": null,
      "cached_tokens": null,
      "text_tokens": 41,
      "image_tokens": null
    }
  },
  "estimated_cost": {
    "unit": "0.0000132500",
    "irt": 1.52,
    "exchange_rate": 114600
  }
}

Gemini 3.1 Flash-Lite via Native v1beta Endpoint

bash

curl https://api.avalai.ir/v1beta/models/gemini-3.1-flash-lite:generateContent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          {"text": "Extract the company names from: Apple announced a partnership with OpenAI and Microsoft this quarter."}
        ]
      }
    ]
  }'

Migration Notes

Existing integrations using gemini-3.1-flash-lite-preview can switch to the stable alias gemini-3.1-flash-lite without code or pricing changes.
The preview alias remains available for backward compatibility.
gemini-3.5-flash is a separate, more capable flagship Flash model — not a direct replacement for gemini-3.1-flash-lite. Choose based on your quality/cost trade-off:
- Use gemini-3.5-flash for agentic, coding, multimodal reasoning, and long-context tasks where quality matters.
- Use gemini-3.1-flash-lite for high-volume classification, extraction, translation, and routing tasks where cost matters most.

New Flagship Gemini 3.5 Flash Added and Gemini 3.1 Flash-Lite Stable Released ​

Summary ​

Details ​

Google (Gemini) ​

Gemini 3.5 Flash ​

Gemini 3.1 Flash-Lite (Stable Release) ​

Pricing Summary ​

API Request/Response Examples ​

Gemini 3.5 Flash Example (OpenAI-Compatible) ​

bash ​

python ​

javascript ​

Gemini 3.5 Flash with Thinking Configuration ​

bash ​

python ​

javascript ​

Gemini 3.5 Flash via Native v1beta Endpoint ​

bash ​

python ​

javascript ​

Gemini 3.1 Flash-Lite Stable Example ​

bash ​

python ​

javascript ​

Gemini 3.1 Flash-Lite via Native v1beta Endpoint ​

Migration Notes ​

Documentation ​

New Flagship Gemini 3.5 Flash Added and Gemini 3.1 Flash-Lite Stable Released

Summary

Details

Google (Gemini)

Gemini 3.5 Flash

Gemini 3.1 Flash-Lite (Stable Release)

Pricing Summary

API Request/Response Examples

Gemini 3.5 Flash Example (OpenAI-Compatible)

bash

python

javascript

Gemini 3.5 Flash with Thinking Configuration

bash

python

javascript

Gemini 3.5 Flash via Native v1beta Endpoint

bash

python

javascript

Gemini 3.1 Flash-Lite Stable Example

bash

python

javascript

Gemini 3.1 Flash-Lite via Native v1beta Endpoint

Migration Notes

Documentation