Name: Gemini 3.8 Flash API
Brand: Google

Overview

What Is Gemini 3.8 Flash API?

Gemini 3.8 Flash, internally codenamed "Skimaki," is Google's most intelligent Flash model, released September 2, 2026 — the third Flash update in about five weeks, following 3.6 Flash (July 21) and 3.7 Flash (August 13). It's engineered for long-horizon software engineering, autonomous agents, and complex enterprise workflows, at Flash speed and cost efficiency. Google also released a cybersecurity-focused sibling, Gemini 3.8 Flash Cyber, alongside it.

Technical Reference

Developer	Google
Model ID	google/gemini-3.8-flash
Context window	1,048,576 tokens
Max output	65,536 tokens
Supported inputs	Image, Text
Output	Text

API Pricing

Input: $0.975 / 1M tokens
Output: $4.875 / 1M tokens
Cached input: $0.0975 / 1M tokens
Google Search grounding: $0.0455 per call

What Gemini 3.8 Flash Brings to the Table

Strong agentic coding benchmarks
Google reports Gemini 3.8 Flash ahead of Anthropic's Opus 5 and OpenAI's GPT-5.6 Sol on several practical benchmarks, including Vals Finance Agent v2 (61.4% vs 58.6% and 53.8%), CharXiv Reasoning (86.2% vs 83.7%), and LVBench agentic mode (87.8% vs 75.4%).

Built for long-running agentic loops
Designed for multi-step orchestration, where an agent breaks a complex task into sequential actions and executes them with less human hand-holding — code generation, refactoring, and production-ready agents are the headline use cases.

Same speed and cost as 3.7 Flash
Google says the coding, agentic, and reasoning upgrades come at the same speed and low cost as the previous Flash release.

Sources: blog.google, 9to5google.com, officechai.com

Uptime and performance

Uptime (24h): 100%
Uptime (7d): 100%
Availability (24h): 99.0%
Error rate (24h): 1.0%
End-to-end latency (24h median): 11.1s
Time to first token (24h median): 5.3s
Throughput (24h median): 119.9 tok/s

Measured on the AI/ML API gateway across live traffic. Figures are medians unless stated. Last updated 2026-10-12.

Use cases

Gemini 3.8 Flash is documented as suited for long-horizon software engineering, including real-world coding benchmarks and complex multi-file refactoring. It is also suited for autonomous agents, with resilient multi-step planning and tool orchestration workflows.

It is documented for complex enterprise workflows, including demanding domain tasks and large-scale data pipelines with deep reasoning and high factual rigor. Google also describes it as well-suited for general-purpose, production-ready agents and agentic knowledge workflows.

It is additionally presented for interactive app building and rich visual experiences, including design adherence, interactive components, screenshots-to-app workflows, and 3D/physics-oriented app creation.

How to use Gemini 3.8 Flash API

Install any OpenAI-compatible SDK, point it at api.aimlapi.com/v1, and set the model to google/gemini-3.8-flash.

import requests

r = requests.post(
    "https://api.aimlapi.com/v1/chat/completions",
    headers={"Authorization": "Bearer " + AIMLAPI_KEY},
    json={
      "model": "google/gemini-3.8-flash",
      "messages": [
        {
          "role": "user",
          "content": "Hello!"
        }
      ]
    },
)
print(r.json())

const r = await fetch("https://api.aimlapi.com/v1/chat/completions", {
  method: "POST",
  headers: {
    Authorization: `Bearer ${process.env.AIMLAPI_KEY}`,
    "Content-Type": "application/json",
  },
  body: JSON.stringify({
    "model": "google/gemini-3.8-flash",
    "messages": [
      {
        "role": "user",
        "content": "Hello!"
      }
    ]
  }),
});
console.log(await r.json());

curl -X POST https://api.aimlapi.com/v1/chat/completions \
  -H "Authorization: Bearer $AIMLAPI_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"google/gemini-3.8-flash","messages":[{"role":"user","content":"Hello!"}]}'

OpenAI-compatible — swap the base URL and it works with your existing SDK.