Language model · by Google

Gemini 3.8 Flashpowers the free tier

Google's newest Flash model reads text, images, video, audio and PDFs — and it's the model behind every free Auto Edit and Clipping render in Shorz.

  • Free tier
  • From 195 credits per 1M tokens (input)
  • Context · 1M
  • Positioning · Free tier

Gemini 3.8 Flash at a glance

Gemini 3.8 Flash is Google's fast multimodal model. Shorz uses it for the free tier: four full Auto Edit or Clipping renders a week for accounts with no credits, where the main AI, transcription and web image search are free. Paid users can pick it as their main AI too.

Specs in Shorz

Made by
Google
Role in Shorz
Main AI; the model for free Auto Edit and Clipping runs
Context window
1,048,576 input tokens
Max output
65,536 tokens
Inputs
Text, image, video, audio, PDF
Released
2 September 2026

Price in Shorz credits

195 / 975credits per 1M tokens · in / out

  • Input · per 1M tokens195 cr
  • Output · per 1M tokens975 cr

1 credit = €0.01. Prepaid packs, no subscription, credits never expire. Prices are read from the same live catalog the app uses. See packs

Why pick Gemini 3.8 Flash in Shorz

Runs the free tier

Every free Auto Edit and Clipping render runs on Gemini 3.8 Flash — no card needed.

Natively multimodal

Reads text, images, video, audio and PDFs.

Low per-token price

Cheaper per token than Sonnet 5.5 and GPT 6.1 Sol.

Best for

  • Free Auto Edit
  • Free Clipping
  • Multimodal analysis
  • Budget projects

How to use Gemini 3.8 Flash in Shorz

Shorz is a desktop app for Windows and Mac. Download it, sign in, and Gemini 3.8 Flash is already in the model picker — no API key, no waitlist, no second subscription.

  1. 1Download Shorz for Windows or Mac and sign in — no card needed.
  2. 2Start an Auto Edit or Clipping project; free runs use Gemini 3.8 Flash automatically.
  3. 3With credits, you can also pick Gemini 3.8 Flash as the main AI for any project.

Gemini 3.8 Flash vs other language models

All language models in Shorz, with live credit prices.

ModelBest atContextPositioningCredits
Claude Opus 5.5Anthropic's Opus 5.5 — the default brain behind every Shorz project.1MDefault520 / 2,600 per 1M tokens · in / out
Claude Fable 5.1Anthropic's flagship Claude 5 model, for the hardest edits and most careful judgement.—Flagship1,300 / 6,500 per 1M tokens · in / out
Claude Sonnet 5.5Anthropic's fast, efficient Claude — half the per-token price of Opus 5.5.1MBalanced260 / 1,300 per 1M tokens · in / out
GPT 6.1 SolOpenAI's GPT 6.1 Sol — 1.05M context, priced alongside Sonnet 5.5.1.05MBalanced270 / 1,350 per 1M tokens · in / out
GPT 6 AstraOpenAI's flagship GPT 6 model — the strongest OpenAI tier in Shorz.1.05MFlagship1,350 / 6,750 per 1M tokens · in / out
GPT 6 LunaOpenAI's smallest GPT 6 model — by far the cheapest main AI in Shorz.1.05MBudget14 / 68 per 1M tokens · in / out
Gemini 3.8 FlashGoogle's multimodal Flash model — and the model behind free Auto Edit and Clipping runs.1MFree tier195 / 975 per 1M tokens · in / out
Grok 4.3 X SearchxAI's Grok 4.3 with live X search — trend research for your AI agent.—Live X search169 / 338 per 1M tokens · in / out

Every model, one app

Shorz isn't a model playground. The models generate the parts; the editor around them turns those parts into a finished, captioned, published video.

  • Every model on one prepaid balance — no per-provider subscriptions
  • No watermark on anything you generate or export
  • Full commercial rights on what you create
  • No API keys to manage: the models run on your Shorz account
  • New model versions appear in the picker without an app update
  • Drivable from Claude, Cursor, Codex or Gemini over MCP

Gemini 3.8 Flash questions

On free Auto Edit and Clipping runs (four full renders a week for accounts with no credits), the main AI is Gemini 3.8 Flash and it costs nothing. Image, video, music, avatar generation and publishing still bill credits.

Yes — pick it in the main AI picker. It bills per token in Shorz credits; live rates are on this page.

Text, images, video, audio and PDF, with a 1,048,576-token input window and up to 65,536 output tokens.

Still have questions? Ask founder