| Spec | Value |
|---|---|
| Maker | Google DeepMind |
| Input | Images, video, audio and text |
| Output | Text; native image and speech output announced |
From Google's December 2024 launch post and its benchmark table, which covers Gemini 2.0 Flash Experimental against the Gemini 1.5 002 models; the post gives no context window or price. · 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 2.0 Flash experimental, 11 Dec 2024). Reasoning null: no Artificial Analysis Intelligence Index figure published at launch that I could open (AA's page now carries a rescaled 8-9); Google's headline is 'outperforms 1.5 Pro on key benchmarks at twice the speed', a relative claim. Coding null. Context 1M (Vals/OpenRouter list 1,048,576) -> 4. Price null: experimental release was free; first paid list price $0.10 in / $0.40 out per 1M (about $0.18 blended, would score 5) appeared with GA in Feb 2025 (Wayback pricing page 15 Feb 2025), outside the launch window. Speed null (Google: twice as fast as 1.5 Pro, no tokens/s). Modalities 5: image, video, audio, text in (native image/audio output announced, early access). Reviews: Simon Willison is hands-on; Chrome Unboxed repeats Google's pitch; SmartWorld (Italian, 31 Jan 2025, stable version) is a side-by-side vs 1.5 Flash.[1]
Given how complicated that photograph is I think this is a pretty amazing result.
2.0 Flash offre una risposta più discorsiva e chiara (prima immagine), mentre 1.5 Flash (seconda immagine) è più schematico.
Keep in mind that this is an early preview, so Google warns that it might not always work perfectly and lacks real-time information access and compatibility with some Gemini features.
| Gemini 2.0 Flash: An outstanding multi-modal LLM with a sci-fi streaming mode · Dec 11, 2024 | 0.70 |
|---|---|
| Gemini 2.0 Flash Experimental lands on Android · Dec 17, 2024 | -0.10 |
| Gemini volta pagina e passa alla versione 2.0 Flash stabile: ecco cosa cambia · Jan 31, 2025 | 0.30 |
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 "Introducing Gemini 2.0: our new AI model for the agentic era" is dated 11 December 2024 and says the experimental 2.0 Flash is available to developers and all Gemini users that day; general availability followed on 5 February 2025.
The Wikipedia article covers the 11 December 2024 announcement of Gemini 2.0 Flash Experimental and its Multimodal Live API, native image and speech output, and the Jules coding agent.
Google's 5 February 2025 post moved 2.0 Flash to general availability and added 2.0 Flash-Lite and an experimental 2.0 Pro.[1]
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.
Epoch AI model database (Gemini 2.0 Flash); Epoch's confidence in the figures: Unknown. Blank tiles are not published.