Today we are very happy to share updates across the board to Gemini 2.5 Model Family:
- Gemini 2.5 Pro is usually available and strong (no changes from 06-05 viewing first)
- Gemini 2.5 Flash is usually available and stabilizing (no changes from 04-17-17 views, see pricing updates below)
- Gemini 2.5 Flash-Lite is now available in view
Gemini Models 2.5 Interested models, can think of their thoughts before answering, resulting in advanced performance and advanced functioning. Each model has a budget management, provided the power enhancements to choose whether the “imaginative” before issuing the answer.
Views of our Gemini Family 2.5 Thinking Models
Introducing Gemini 2.5 Flash-Lite
Today, we introduce 2,5 flash-lite by viewing the lowest latency and costs in a 2.5 model family. Designed as negative development in our 1.5 and 2,5-flash models. It also provides the best working throughout the Evels, and the lowest time in the first sign when reaching higher tokens for the second progress. This model is good for high-ending tasks as parting or summarizing.
Gemini 2.5 Flash-Lite is a consulting model, allowing a strong budget management of the API parameter. Because Flash-Lite is made of cost and speed, “thinking” is automatically closed, unlike other models of our. 2.5 Flash-Lite and supports all our traditional tools such as Google placement, Code, and URL context in addition to operation.
Gemini Company Backs 2.5 Flash-Lite
Update of Gemini 2.5 Flash and prices
Last year, our research groups continue to pressure the pareto border with our Flash Model chain. When A 2.5 Flash was originally announced, we had not completed 2,5 Flash-Lite skills. We also started “thinking” and “unwanted price”, which led to engineers.
With a settlement version of Gemini 2.5 Flash Rolling Out (which is the same 05-20 model view that made it available at Google I / O) 2,5 flash, we update the price of 2.5 Flash prices.
- Entering tokens for $ 0.30 / 1M (* up from $ 0.15 Input)
- Relaxing tokens $ 2.50 / 1m (* down from $ 3.50 output)
- We have removed thoughts vs vs vs differing price difference
- We kept a single tier of price no matter what the installation token size
While we strive to maintain unchangeable prices between priorities and releases to reduce the disturbance, this is something to show a different flash value, which provides the best cost of Intelligence.
And with a Femino 2.5 Flash-Lite, we now have a very low choice (thinking or without thinking) at cost and cases critical use of latency need a small model intelligence.
Price updates in our family Famin Flash family
If you use Gemini 2.5 Flash Preview 07-17, the first viewing amount will remain there until July 15, 2025, where it will be extinguished. You can turn into converted model “Gemini-2.5-flash” model, or switch to 2,5 flash-lite screen as a low cost option.
Continuous Growth Green 2.5 Pro
The growth and need for a gasignment of Gemini 2.5 Pro continues to have some of our models that you have ever seen. Allowing a lot of customers that build in this model in production, we make a 06-05 version stable model, with the same point of pero proloro point as before.
We expect the cases when you need a great deal of wisdom and so much energy where you will see a SYRONE, like Conting and Agentic Activities. Gemini 2.5 Pro is in the heart of many popular tools for engineers.
Top Engineer Tools use Gemini 2.5 Pro
If you are using 2.5 Pro Preview 05-06, the model will remain available until June 19, 2025 and turn off. If you are using 2.5 Pro Preview 06-05, you can simply renew your model strand at “Gemini-2.5-Pro”.
We will not wait to see more domains benefit from 2,5 Pro intelligence and look forward to sharing more in comparison more than pro.
