Models & pricing / gemini-3.5-flash-lite · Google
Google

gemini-3.5-flash-lite API

Gemini 3.5 Flash-Lite is Google’s multimodal model for frequent subtasks, document parsing and simple extraction. It produces text from questions about images, audio, video or PDF material. On XiuAI, compare task quality, current cost and the selected API protocol.[1][2]

gemini-3.5-flash-lite

Model overview

Published provider specifications. Available features depend on the current API route.

API routes
Gemini GenerateContent / OpenAI-compatible API / openai-response / openai-response-compact / Anthropic Messages / openai-alpha-search

Defining high-volume subtasks

Evaluate it for classification, field extraction, document triage and bounded agent subtasks. Give each request a clear input and output contract, such as extracting invoice fields or judging relevance. A cost-focused position is not a reason to assign every complex reasoning task to it.[1]

Google positions Flash-Lite for latency- and API-cost-sensitive workflows. Actual turnaround here still depends on input, output and routing. Evaluate errors and the cost of manual correction on your own samples before scaling.[1]

Multimodal input and text output

The official model page lists text, image, video, audio and PDF input, with text output. Reading a recording or video does not imply speech synthesis, video generation or Live API support. Encode media and files for the API you actually use.[1]

Comparing the cost of completing a task

Use the current input, output and cache rates shown here. For frequent extraction, avoid repeatedly sending unnecessary material or generating verbose answers, and track retries and manual corrections. The vendor’s low-cost positioning is not a lowest-price guarantee for every XiuAI request.[2][1]

API access

Using Flash-Lite through XiuAI

Use gemini-3.5-flash-lite with your existing XiuAI API key. Native Gemini requests use contents and parts; compatibility APIs use different message, media and tool formats. Convert to the selected API rather than changing only the URL, and validate one required input type before adding others.[2][1]

Send a request using your existing API key.

POST /v1/chat/completions
curl 'https://router-api.xiu.ai/v1/chat/completions' \
  -H "Authorization: Bearer $XIUAI_API_KEY" \
  -H 'Content-Type: application/json' \
  --data-raw '{
  "model": "gemini-3.5-flash-lite",
  "messages": [
    {
      "role": "user",
      "content": "Hello"
    }
  ]
}'
Read the API documentation

Frequently asked questions

Should Flash-Lite be the default for every task?

Its vendor positioning emphasizes high-throughput, low-cost subtasks and document parsing. Validate bounded extraction or classification first. Compare with suitable catalog models for complex planning or multi-step reasoning.[1]

Does audio or video input mean it can generate media?

No. Its output is text, and the official page does not list audio generation, image generation or Live API support. Choose an available generation model when you need media output.[1]

Can I directly use Google-hosted search and execution?

Those are vendor platform tools, not capabilities established for the current XiuAI route by the model description. Check the API and fields separately; application tools, model function calls and vendor-hosted tools are different layers.[1][2]

Related models

gemini-3.8-flashgemini-4-hydrogen

Sources and updates

Specifications reference provider documentation; prices and availability come from the current XiuAI catalog.

Content reviewed: 2026-11-16

[1] Google Gemini 3.5 Flash-Lite documentation[2] Current XiuAI model and pricing dataModel documentationTool callingStructured outputContext documentation