| Spec | Value |
|---|---|
| Maker | Google DeepMind |
| Context window | 1M tokens |
| API price per 1M tokens (input / output) | $0.10 / $0.40 (audio input $0.50) |
From Google's launch posts and the benchmark and pricing images on the developer blog (preview 06-17 pricing). Gemini scores are pass@1 with default settings; the pages state no maximum output. · source
Estimate from the version's published specs, benchmarks and reviews, 0-5; scores are only comparable within one criteria year. Scored as launched (Gemini 2.5 Flash-Lite preview, 17 Jun 2025). Reasoning null: no launch-scale AA index figure (AA's page now shows 7 non-thinking / 9 thinking on its rescaled index); Google gave only relative claims (higher quality than 2.0 Flash-Lite). Coding null. Context 1M -> 4 (Google post). Price 5: preview list price $0.10 in / $0.40 out per 1M (Google pricing page, Wayback 18 Jun 2025; The Decoder), about $0.18 blended, same with thinking. Speed null: Google claimed lower latency than 2.0 Flash-Lite and 2.0 Flash, no tokens/s at launch (AA's later 295-377 t/s is a later measurement, not used). Modalities 5. Reviews: only one hands-on (Simon Willison, found it looped on an audio transcript); The Decoder and heise are launch reports.[1]
but the new Flash Lite model got stuck in a loop (65,517 output tokens = 6.3241 cents , 231.9 seconds) part way into the transcript
Flash-Lite delivers low prices and fast response times, while Pro leads in complex reasoning and accuracy.
Google claims to have improved the quality of Gemini 2.5 Flash-Lite in many areas such as coding, mathematics, science, reasoning and multimodal benchmarks.
| Trying out the new Gemini 2.5 model family · Jun 17, 2025 | -0.40 |
|---|---|
| Gemini 2.5 Flash-Lite is the fastest and most cost-effective model in Google's Gemini lineup · Jun 17, 2025 | 0.30 |
| Gemini 2.5 Flash-Lite is Google's fastest and most cost-effective AI model · Jun 18, 2025 | 0.40 |
Photos from readers, kept apart from the pin's sourced media. Rate what you ate or saw, and on a restaurant pin tag the dishes right on your photo.
The first entry is always the pin's source. Overall confidence is a weighted average of how firmly each reference supports the start and end times used above; a reference counts half as much for every 180 days older than the newest.
Google's post "We're expanding our Gemini 2.5 family of models" is dated 17 June 2025 and opens a preview of 2.5 Flash-Lite in Google AI Studio and Vertex AI that day.
The Wikipedia article lists Gemini 2.5 Flash-Lite as released on 17 June 2025.
Something missing or wrong? Say it in your own words: a link that backs this pin up, a different start or end date and why, or a fact it lacks or gets wrong. The AI checks it against this pin's sources, searches for better ones, and adds any page that backs you up. The pin's own sources still count most. A picture that shows something else, or shows it badly, is looked at too, and moved down or replaced.