Aug 11, 2024 13:00:00

Google's multimodal AI 'Gemini 1.5 Flash' usage fees cut by up to 78%

The AI model '

Gemini 1.5 Flash ' announced by Google in May 2024 is characterized by its low price, which is only one-tenth of the price of Gemini 1.5 Pro, despite its performance being comparable to the high-performance Gemini 1.5 Pro in benchmark tests. Google has also announced that it will implement a significant price reduction for Gemini 1.5 Flash from August 12, 2024.

Gemini 1.5 Flash price drop with tuning rollout complete, and more - Google Developers Blog
https://developers.googleblog.com/en/gemini-15-flash-updates-google-ai-studio-gemini-api/

Gemini API Pricing | Google AI for Developers

https://ai.google.dev/pricing

Good news for @GoogleAI developers:

- Gemini 1.5 Flash price is now ~70% lower ($0.075 / 1M)
- Gemini 1.5 Flash tuning available to all
- Added support for 100+ new languages in the API
- AI Studio is available to all workspace customers
- Much more : ) https://t.co/LmN4KiDBWq

— Logan Kilpatrick (@OfficialLoganK) August 8, 2024

The Gemini 1.5 Flash, a lightweight model with performance comparable to the Gemini 1.5 Pro, has a context window of 32,000 tokens, which is four times that of the Gemini 1.0 Pro, and can process a lot of information at once. For more information on Gemini 1.5 Flash, see the following article.

Google announces 'Gemini Flash', a high-speed, high-performance lightweight AI model, with performance equivalent to that of Gemini Pro at one-tenth the price - GIGAZINE

Gemini 1.5 Flash was previously available for free , but Google also offered paid plans with relaxed rate limits. Now, Google has announced that it will lower the price of these plans.

Below is a table showing the fee structure before and after the price reduction per 1 million tokens. Below 1.28 million tokens, a significant price reduction of more than 70% is planned for both input and output.

	Number of tokens	input		output	Context Cache
Before price reduction	Greater than 128,000	$0.70 (about 102 yen)		2.10 dollars (about 308 yen)	$0.175 (approx. 25 yen)
After price reduction		$0.15 (about 22 yen)		0.6 dollars (about 88 yen)	0.0375 dollars (approximately 5.5 yen)
Before price reduction	Under 128,000	$0.35 (about 51 yen)		1.05 dollars (about 154 yen)	0.0875 dollars (approximately 12 yen)
After price reduction		0.075 dollars (about 11 yen)		0.3 dollars (about 44 yen)	0.01875 dollars (approximately 2.75 yen)

'With these reduced fees and tools like context caching, developers should see significant cost savings when building systems using Gemini 1.5 Flash's long context and multimodal capabilities,' Google said.

Google also reports that both Gemini 1.5 Pro and Gemini 1.5 Flash now support over 100 languages, and that text tuning has been performed on Gemini 1.5 Flash, which is expected to reduce the context size of prompts, reduce latency, lower costs, and improve model accuracy for the task.

In addition, it was announced that the developer documentation for the Gemini API has been made easier to use and that PDF processing is now possible with the Gemini API.