Gemini 3.8 Flash
google/gemini-3.8-flash
Google's most capable Flash model, with gains over 3.7 Flash in software engineering, agentic tasks, and multi-step reasoning. Takes text, images, video, audio, and files.
RoleThe cheap multimodal worker
$0.825
In / M
$4.13
Out / M
1.0M
Context
- AA intelligence index
- 30.8
- Inputs
- text, image, video, file, audio
- Outputs
- text
- Tools
- Listed
- Structured output
- Listed
- Reasoning
- Listed
- Weights
- Not verified
- Listed
- Sep 2, 2026
01 · What it's good at
Use it for this.
Three jobs this model is built for, and the ones to give to something else.
01
Cheap multimodal: audio, video, images, and PDFs in one call
02
Agent steps and coding at Flash pricing
03
Long files thanks to a 1M-token context
Skip it for
- Your hardest agent turns. Gemini 4 Argon is the flagship
02 · Example tasks
What you'd actually send it.
Real inputs, not a benchmark. Paste one into the snippet below to try it.
You send
From this 30-minute meeting recording, list decisions, owners, and due dates as a table.
Why this model
Audio in, structured text out, at a Flash price.
03 · Your stack
Where it sits: Do the volume.
The worker for media-heavy pipelines: per upload, per recording, per screenshot. Escalate the unclear cases to Argon.
Layer 01
Retrieve
Layer 02
Plan & reason
Layer 03 · this model
Do the volume
Layer 04
Gate & route
Layer 05
Draw, speak, listen
04 · Wire it up
One request. Paste and run.
Uses OpenRouter, so one key works for every model on this site. Set OPENROUTER_API_KEY and swap the prompt for one of the tasks above.
// OpenRouter speaks the OpenAI Chat Completions format.
const res = await fetch("https://openrouter.ai/api/v1/chat/completions", {
method: "POST",
headers: {
Authorization: `Bearer ${process.env.OPENROUTER_API_KEY}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"model": "google/gemini-3.8-flash",
"messages": [
{
"role": "user",
"content": [
{
"type": "text",
"text": "Read this receipt and return merchant, date, total, and currency."
},
{
"type": "image_url",
"image_url": {
"url": "https://example.com/receipt.jpg"
}
}
]
}
],
"reasoning": {
"effort": "low"
}
}),
});
const data = await res.json();
console.log(data.choices[0].message.content);
05 · Watch out for
Gotchas.
Reasoning is mandatory on the listing (low, medium, high). Use low for extraction.
Batch pricing is listed separately for offline jobs.
Sources
Lessons are written from each model's listing and the linked docs. We don't publish our own benchmark numbers; the index above is Artificial Analysis, as shown on OpenRouter.
Keep learning