What happened
On July 21, 2026, Google released two lower-cost models in its Gemini line, Gemini 3.6 Flash and Gemini 3.5 Flash-Lite, both aimed at running AI agents on high volume without the token costs piling up. Google positions Flash for coding and multimodal work and Flash-Lite for fast, high-throughput jobs like document processing and agentic search, with reported per-token prices well below its flagship tier. The pitch is simple: as businesses move from one-off AI questions to agents that run continuously, the speed and cost of the workhorse model is what decides whether that is affordable.
Reported by AI News. Our analysis is below.
The EMOR AI Take
The pattern worth watching is not which lab is ahead this week, it is the direction of the price line, and it points one way: down. Every few weeks another cheaper, faster workhorse model ships, which means the automations we run for clients, lead follow-up, document handling, content, get cheaper to operate over time without anyone changing a thing. We route each job to the model that fits it, so falling prices reach your bottom line instead of a lab’s margin.
There is a reason this matters for a small business specifically. The falling cost of these workhorse models is what turns AI from a one-off novelty into something you can run on real volume, every lead, every document, every day, at a price that still makes sense. The businesses that build that into their operations now compound the advantage as the underlying cost keeps dropping. Put cheap, reliable intelligence to work on your routine volume and let the market keep lowering your bill.
What this means for your business
- Judge AI vendors on whether they pass falling model prices to you
- Move routine, high-volume work onto automation now
- Expect the cost of running AI agents to keep dropping
Read the original report
AI News
Want this working for your business?
We build the AI receptionists, automations, and content systems businesses actually run on, custom or product.
