Gemini 3.8 Flash

API Model ID:gemini-3.8-flash

Real-time multimodal speed demon with 1M context and near-instant time-to-first-token.

#Sub-70ms#1M Context#Fast Multimodal#Real-time Streaming#Vision

Deploy with Gruvo All Models

Context Window1M tokens1,000,000 tokens

Max Output32K tokens32,000 tokens

Input Price$0.15per 1M input tokens

Output Price$0.60per 1M output tokens

Latency / Speed~65ms260 tokens/sec

Overview & Architecture

Gemini 3.8 Flash delivers Google's fastest multimodal inference. With sub-70ms latency, high tokens-per-second streaming, and a 1-million-token context window, 3.8 Flash is designed for real-time human-in-the-loop applications, streaming video processing, and high-frequency automated tools.

ArchitectureDistilled Next-Gen Multimodal MoE with Flash Attention V4

Knowledge IndexCurrent (Continuously Indexed)

Benchmark Highlights

Latency TTFT65ms

Ultra-fast first-token response

MMLU86.9%

General knowledge benchmark

MathVista74.8%

Visual mathematical reasoning

Supported Capabilities

Function Calling / Tools

Supported

Vision & Image Inputs

Supported

Audio & Voice Inputs

Supported

Structured JSON Mode

Supported

System Prompts

Supported

Streaming Completions

Supported

Prompt Caching

Supported

Engineering Strengths

Ultra-fast 65ms TTFT and blazing 260 tokens/sec sustained throughput
Massive 1M token context capacity at budget pricing
Full multimodal capability: processes voice, images, and video on the fly
Generous rate limits and rapid horizontal scaling

Recommended Production Workloads

Interactive real-time voice and video agents with streaming responses
High-throughput RAG search pipelines indexing multi-document corpora
Rapid visual inspection and instant screen-share diagnostics

Execute via Gruvo

Unified Endpoint

Connect through Gruvo with OpenAI-compatible client libraries. Gruvo translates requests, manages provider streaming, and applies caching automatically.

import OpenAI from "openai";

// Configure client with Gruvo's unified AI gateway
const gruvo = new OpenAI({
  baseURL: "https://api.gruvo.ai/v1",
  apiKey: process.env.GRUVO_API_KEY,
  defaultHeaders: {
    "X-Gruvo-Provider": "gemini",
  },
});

const completion = await gruvo.chat.completions.create({
  model: "gemini-3.8-flash",
  messages: [
    { role: "system", content: "You are a production reasoning assistant." },
    { role: "user", content: "Analyze the architecture of our service." },
  ],
  temperature: 0.2,
});

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