New Flagship Model Added: Gemini 3.7 Flash

Date: 2026-08-14 / (1405-05-23)

Summary

Google's new flagship Flash model, gemini-3.7-flash, is now available through AvalAI for coding, agents, web development, document analysis, and complex knowledge work. It supports the native Gemini v1beta/, v1/chat/completions, and v1/messages, with partial support for v1/responses.

---

Details

Google Gemini 3.7 Flash

gemini-3.7-flash is Google's most capable Flash workhorse model for coding and agentic workflows. It builds on Gemini 3.6 Flash with improved software engineering, web development, complex document comprehension, business workflow automation, instruction following, multi-step planning, and tool use.

Key Features:

• Long Context: 1,048,576 input tokens and up to 65,536 output tokens
• Multimodal Input: Text, image, video, audio, and PDF input with text output
• Coding and Agents: Improved debugging, issue resolution, long-horizon software engineering, planning, and tool calls
• Web Development: Better first-pass code accuracy, design adherence, and feature-complete application generation
• Knowledge Work: Stronger reasoning over complex documents in finance, law, biosciences, and other knowledge-dense fields
• Workflow Automation: Improved completion of multi-skill business workflows with less manual oversight and fewer retries
• Developer Features: Thinking, function calling, structured outputs, code execution, prompt caching, file search, Google Search grounding, and URL context
• Best For: Agentic coding, production application development, document intelligence, workflow automation, and complex multimodal tasks

Benchmark Highlights Reported by Google

Compared with Gemini 3.6 Flash, Google reports the following results:

Evaluation	Gemini 3.7 Flash	Gemini 3.6 Flash
FrontierCode 1.1 Main	43.6%	34.4%
DeepSWE v1.1	65.3%	49.0%
WebDev Arena	1588 Elo	1538 Elo
GDP.pdf	34.0%	22.0%
AutomationBench	30.4%	17.0%

Benchmark results are useful directional evidence, but production evaluation should use prompts, tools, and acceptance criteria representative of your own workload.

Endpoint Availability

Endpoint	Support	Notes
v1beta/	Supported	Native Gemini request and response schema
v1/chat/completions	Supported	OpenAI-compatible Chat Completions
v1/messages	Supported	Anthropic-compatible Messages API
v1/responses	Partial support	Verify required parameters and tools before production use

---

Promotional Pricing

Prices are in USD per 1 million tokens. Google's introductory pricing applies through December 31, 2026.

Period	Input	Cached Input	Output
Through December 31, 2026	$0.75	$0.075	$3.75
After December 31, 2026	$1.50	$0.15	$7.50

The promotional rates are half the standard rates. Review the pricing page before deploying long-running workloads whose traffic may continue beyond the promotional period.

---

API Request and Response Example

Request

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "gemini-3.7-flash",
    "messages": [
      {
        "role": "user",
        "content": "Review this service design and propose a phased migration to an event-driven architecture. Include risks and rollback criteria."
      }
    ]
  }'

Response

The following shortened response illustrates the standard Chat Completions structure. Token counts and cost vary with the request and generated output.

{
  "id": "chatcmpl-gemini37-example",
  "created": 1786651200,
  "model": "gemini-3.7-flash",
  "object": "chat.completion",
  "system_fingerprint": null,
  "choices": [
    {
      "finish_reason": "stop",
      "index": 0,
      "message": {
        "content": "Begin with domain and dependency mapping, then introduce versioned events and a transactional outbox behind existing service contracts. Migrate one low-risk domain at a time, measure dual-write consistency, and retain a traffic-switch rollback path until consumer lag and error-rate targets remain stable.",
        "role": "assistant",
        "thinking_blocks": [],
        "annotations": []
      }
    }
  ],
  "usage": {
    "completion_tokens": 160,
    "prompt_tokens": 24,
    "total_tokens": 184,
    "completion_tokens_details": null,
    "prompt_tokens_details": {
      "audio_tokens": null,
      "cached_tokens": null,
      "text_tokens": 24,
      "image_tokens": null
    }
  },
  "estimated_cost": {
    "unit": "0.0006180000",
    "irt": 70.82,
    "exchange_rate": 114600
  }
}

---

SDK Usage Examples

Bash

curl https://api.avalai.ir/v1/chat/completions \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "model": "gemini-3.7-flash",
    "messages": [
      {
        "role": "user",
        "content": "Find the reliability risks in this deployment plan and return a prioritized mitigation checklist."
      }
    ]
  }'

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-3.7-flash",
    messages=[
        {
            "role": "user",
            "content": "Find the reliability risks in this deployment plan and return a prioritized mitigation checklist.",
        }
    ],
)

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-3.7-flash",
  messages: [
    {
      role: "user",
      content: "Find the reliability risks in this deployment plan and return a prioritized mitigation checklist.",
    },
  ],
});

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

Native Gemini API Example

Use the native Gemini endpoint when your application needs Gemini-specific request fields or tools:

curl https://api.avalai.ir/v1beta/models/gemini-3.7-flash:generateContent \
  -H "Content-Type: application/json" \
  -H "Authorization: Bearer $AVALAI_API_KEY" \
  -d '{
    "contents": [
      {
        "role": "user",
        "parts": [
          {
            "text": "Design a tool-using agent that triages incidents and prepares a human-reviewable remediation plan."
          }
        ]
      }
    ]
  }'

---

Migration Guidance

• Use the exact model ID gemini-3.7-flash when adopting the new model.
• Evaluate the model against representative coding, tool-use, and document workloads before changing production traffic.
• Existing Gemini 3.6 Flash applications can generally retain their endpoint and request structure while changing the model ID.
• Confirm support for every required parameter and tool before using the partially supported v1/responses endpoint.
• Account for the standard rates that take effect after December 31, 2026 when forecasting long-term cost.

---

Related Links

• Google Models Documentation
• Gemini 3.7 Flash Model Reference
• Chat Completions API Reference
• Native Gemini API Reference
• Pricing