Video: "Google AI Studio + Gemini 3.5 Flash is INSANE!" by Julian Goldie on YouTube.
What Gemini 3.5 Flash actually is
Gemini 3.5 Flash is Google's speed-optimised model from the 3.5 family announced at I/O 2026. The headline numbers: frontier-tier performance at roughly four times the throughput of comparable models, at $1.50 per million input tokens and $9 per million output tokens. Benchmark performance sits at 76.2% on Terminal-Bench 2.1, which puts it in the same bracket as models costing considerably more to run.
The emphasis on speed is deliberate. When an AI model is powering a multi-step agent workflow — keyword research, then content clustering, then page drafting — the latency of each step compounds. A faster base model means the whole pipeline completes in minutes rather than tens of minutes. That is the practical reason Google positioned this as an agent-first model rather than a general chatbot upgrade.
Why Google AI Studio changes the access question
Previously, serious use of Gemini models at volume meant navigating Google Cloud Platform — project setup, billing accounts, IAM permissions, Vertex AI endpoints. That is fine if you have a GCP environment already, but it is a real barrier for a small team that just wants to build a content tool or run an SEO workflow.
Google AI Studio is the lighter-weight route: a browser-based interface with direct API access, a prompt builder, and a code generator that produces Python or JavaScript for whatever you build. You get a free tier to start, and upgrading to pay-per-use does not require a full GCP account. Julian Goldie used it to build an SEO keyword planning tool without writing backend infrastructure from scratch.
What this looks like for SEO work
The combination Julian Goldie demonstrated covers the repetitive middle of an SEO workflow. Feed it a topic and a list of seed keywords and it clusters them by intent, suggests supporting page structures, and drafts meta descriptions at volume. The speed of 3.5 Flash means you can iterate quickly — adjust the brief, rerun, review — rather than waiting minutes between outputs.
Worth knowing: the model has a one million token context window. That is large enough to pass in an entire site's existing content alongside a new brief and ask for gap analysis or internal link suggestions. Most SEO tools do not offer that in a single prompt.
The honest limits
Gemini 3.5 Flash is optimised for speed and agent throughput, not for the kind of nuanced long-form reasoning that Claude Opus delivers. For bulk content production it is well-suited. For detailed topical authority mapping, brand voice work, or anything requiring tight editorial control, a slower and more careful model will still produce better output.
Google AI Studio is also not a fully managed environment. It is a development tool. What Julian built in the video is a working prototype, not a production SEO pipeline — and turning it into one would need additional work around error handling, scheduling, and output review.
Where this connects to NordSys
We build AI SEO systems that actually run reliably — keyword research, content briefs, and on-page work handled by the right models for each task. Gemini 3.5 Flash is a strong choice for the high-volume, fast-iteration parts of a search content workflow, and we know where it fits alongside Claude and other tools. If you want a working setup rather than a demo, that is what we do.
See our SEO & AI Ranking service →