Yes. Start with free quota on iMini Agent; after that, continue on a subscription. No separate Google API key required.

Google · Gemini 3.5 Flash · May 2026

Gemini 3.5 Flash — near-Pro ability at a faster pace

Q: Is the name on this page the same model I pick in Agent?

Yes. Choosing Gemini 3.5 Flash in Agent is the model this page describes.

Q: What are the main specs?

1M-token context, max output similar to the Flash family (~64K–65K), adjustable thinking levels, and multimodal input.

Q: Should I move up to Gemini 3.6 Flash?

For new similar work, evaluate 3.6 first. If prompts and evals are already stable on 3.5, stay—and compare on representative tasks before migrating.

Q: How do I choose among Gemini 3.6 Flash and Gemini 3.1 Pro?

Keep proven 3.5 workflows. Want faster and leaner? Look at 3.6. Need higher precision? Use 3.1 Pro.

Q: Can I use the output commercially?

Yes. Content you create on iMini carries commercial use rights under the product terms.

Million-token context
Multimodal input
Adjustable thinking depth
Parallel agent loops

What is Gemini 3.5 Flash?

Gemini 3.5 Flash is Google’s May 2026 release aimed at near-Pro coding and reasoning while keeping faster replies. Versus earlier Gemini 3 Flash Preview, agent loops and thinking controls are more complete. For new similar work, evaluate Gemini 3.6 Flash first; for a higher ceiling, look at Gemini 3.1 Pro. On iMini Agent, pick Gemini 3.5 Flash to use it.

Vendor: Google
Released: May 2026
Context: 1M tokens
Max output: ~64K–65K
Deep thinking: Usually medium · adjustable
Best as: Proven workflows

What’s new versus Gemini 3 Flash Preview

Near-Pro coding, thinking levels, and multimodal agents—the capabilities to check before you stay on 3.5.

Coding and reasoning closer to Pro

Google positions 3.5 Flash as near-Pro coding and reasoning while keeping faster replies and a leaner cost posture.

More complete thinking depth

Tune from lighter to deeper thinking by task; medium depth usually works for both everyday and slightly harder jobs on one model.

Parallel agent loops

Better suited to multi-tool, multi-step workflows that run in parallel—and can keep history in long context.

Full multimodal input

Text, images, video, audio, and PDFs can enter the same task—suited to mixed UI, document, and AV materials.

Official evaluations

Artificial Analysis results cited in Google’s Gemini 3.5 launch materials: intelligence index and output speed on one scatter plot—showing where Flash sits between ability and throughput.

Artificial Analysis: intelligence index vs output speed, highlighting Gemini 3.5 Flash — Artificial Analysis intelligence index vs output speed (data as of 2026-05-13). Gemini 3.5 Flash scores ~57 on the index at ~285 tokens/s—same intelligence band as Claude Opus 4.7 and Gemini 3.1 Pro, with clearly higher throughput. That is the core evidence for the Flash positioning.

Three common workflows

Gemini 3.5 Flash still fits proven workflows; for new picks, compare 3.6.

Near-Pro everyday coding

For feature work and medium-difficulty fixes. Stay on 3.5 if the workflow is locked; try 3.6 for new work.

Long-context agents

For automation that carries long specs and tool traces. Agree on completion criteria up front.

Multimodal material organization

For understanding and summarizing screenshots, PDFs, and AV materials.

How to choose vs Gemini 3.6 Flash and Gemini 3.1 Pro

All three are Gemini 3. The gap is mainly generation efficiency and the Pro ceiling.

Criterion	Gemini 3.5 Flash	Gemini 3.6 Flash	Gemini 3.1 Pro
Context	1M tokens	1M tokens	1M tokens
Reasoning	Usually medium · adjustable	Adjustable thinking, leaner output	Pro-tier reasoning ceiling
Coding & agents	Near-Pro fast coding	Faster, leaner coding	Heavier agents & complex workflows
Long-horizon tools	Parallel agent loops	High-turnaround, more restrained	Favors reliability
Speed posture	Faster replies	Faster and leaner	Favors precision
Prefer when	Workflows still locked on 3.5	New fast coding & agent work	You need higher precision

Using Gemini 3.5 Flash on iMini

Free quota, model naming, key specs, and how it differs from 3.6 and Pro.

Can I try it free?

Is the name on this page the same model I pick in Agent?

What are the main specs?

Should I move up to Gemini 3.6 Flash?

How do I choose among Gemini 3.6 Flash and Gemini 3.1 Pro?

Can I use the output commercially?

Use Gemini 3.5 Flash free on iMini Agent

Million-token context, near-Pro ability, faster replies.

Open Gemini 3.5 Flash

No Google API key required.