Google models
All 54 Google models on OpenDunes, priced in Algerian Dinars.
54 models
This model is designed to handle text, pictures, and video inputs, allowing it to read up to 130 pages of content. It specializes in generating and editing video clips, producing outputs that include both text and moving images. What sets this model apart is its native ability to process and generate video content directly. It is not suitable for tasks requiring external tool usage or the creation of structured data formats.
#1 for speech · Gemini 3.8 Flash TTS turns text into speech in 130+ languages with studio-grade acting, regional accents and a voice that stays consistent through long dialogues. It offers prebuilt voices, custom voice personas and voice replication, with about 8 pages of text per request. Pick it for audiobooks, narration and multi-speaker scenes.
Nano Banana 2, Google's 3.1 Flash Image model, generates and edits images from a written description and reference pictures, returning images with text. It outputs up to 4K, adds very tall and very wide shapes such as 1:8, and renders text in other languages better. Pick it for production batches of ads, banners and edits.
This model accepts text, pictures, and videos as input, allowing it to read up to 130 pages of content. It specializes in creating and editing video content, converting descriptions or pictures into moving images. A key feature is its ability to search the web for information while processing complex visual data. However, it does not utilize external tools or generate structured data formats.
Google's synchronous speech-to-text model for when timestamps on every word or labels for up to eight speakers are needed, on audio of up to an hour, or half an hour with those extras switched on
Flash Lite TTS from the 3.8 family turns text into speech in 101 languages, built for large volumes of audio, live voice assistants and read-aloud features. It offers prebuilt voices, custom voice personas and voice replication, with about 8 pages of text per request. Pick it for speed and volume.
The long-form half of Google's Lyria 3 music models, turning a text prompt or an image into complete songs with verses, choruses and bridges in 48 kHz stereo, with vocals and timed lyrics; Lyria 3 Clip makes loops
Google's preview of 3.1 Flash TTS turns text into speech using Google's prebuilt voices. It is the earlier generation of Gemini speech: Google now points users to Gemini 3.8 Flash TTS for acted-out expression and long multi-turn dialogue, and to 3.8 Flash-Lite TTS for volume. Keep this one only for workflows that write style directions into the text, which the 3.8 models expect as separate settings.
Nano Banana 2.1 generates and edits images from a written description and up to 14 reference images, keeping up to four characters and ten objects consistent across edits. It outputs 1K, 2K or 4K, renders text more cleanly than Nano Banana 2, and can base a picture on web and image search results. Pick it for ad and product visuals that go through many edits.
The short-form half of Google's Lyria 3 music models, turning a text prompt or an image into clips, loops and previews in 48 kHz stereo with vocals, timed lyrics and full arrangements; Lyria 3 Pro makes songs
Nano Banana 2 Lite, Google's quickest Gemini image model, returning a picture in about 4 seconds, 2.7 times faster than Nano Banana 2, while keeping its character consistency, precise edits and world knowledge
Gemini Embedding 2 turns text, images, video, audio and PDFs into fingerprints of their meaning in one shared space, up to about 8 pages per item. That lets one search find a picture from a sentence or a clip from a document. Pick it for search across kinds of media. For text-only collections of pieces under 2 pages, Gemini Embedding 001 is enough.
Gemini Embedding 001 turns text into a fingerprint of its meaning, so a search finds text that means the same thing and not only the same words; it also helps sort text into categories. It takes text only, about 2 pages per piece. Pick it when your collection is plain text. For images, audio, video or PDFs in the same search, and for pieces up to about 8 pages, use Gemini Embedding 2.
#10 for pictures · Image generation and editing with contextual understanding, including edits carried over several turns of a conversation and up to three reference images, in ten shapes; the original Nano Banana, generally available
Nano Banana Pro, or Gemini 3 Pro Image, creates and edits pictures from a written description and reference images, returning images with text. It reasons before drawing and can base a picture on search results, so it fits infographics, product mockups and layouts with exact lettering, up to 4K. Pick it when text and structure must be right. Nano Banana 2 and 2.1 are the faster tier for routine batches.
Chirp 3 is the speech-to-text model from Google: send audio, get a transcript with punctuation added and the language detected for you. It covers 50+ languages and handles clips under a minute and recordings up to an hour, and it can tell speakers apart when you process recordings in batches. Pick it for multilingual call or meeting audio.
The preview build of Google's first multimodal embedding model, placing text and images in one vector space so a written query can find matching pictures and the reverse, with 128 to 3,072 dimensions
Veo 3.1 turns a written description or a picture into video with synchronized sound, aimed at final cuts up to 4K with complex camera moves. It accepts a description of up to about a page and returns one video per request, and it can extend a clip. Pick it when finished quality matters. Veo 3.1 Fast is the quicker mid-tier, and Veo 3.1 Lite suits volume.
#3 for photo editing · This is the preview version of Nano Banana Pro, Google's image model for complex layouts, product mockups and pictures where the lettering must be exact. It takes a written description and reference images and returns images with text, holding about 65 pages of input. Nano Banana 2 or 2.1 are the faster tier for routine batches.
Veo 3.1 Fast turns a written description or a picture into video with synchronized sound, balancing speed against quality. It returns results faster than Veo 3.1 while keeping the same kind of input and one video per request. Pick it for drafts, iterations and social clips where waiting matters.
The lightest Veo 3.1, meant for content platforms and automated media at volume, producing 4 to 8 second clips at 720p or 1080p in landscape or portrait from a prompt or an image, sound included
#4 for pictures · This is the preview version of Nano Banana 2, which generates and edits images from a written description and reference pictures and returns images with text. It holds about 65 pages of input, half of the released 3.1 Flash Image, so pick the released one for new work. For the most precise lettering, step up to Nano Banana Pro.
#5 for games · Gemini 3.8 Flash takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text up to 65 pages long, with low, medium and high thinking levels. Google rates it the strongest Flash model for long software-engineering runs, assistants that act on their own and business workflows.
#1 for games · Gemini 4 Argon takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind, and answers in text up to about 1,000 pages long, with low, medium and high thinking. Google calls it its top model for reasoning. Pick it for complex software engineering, hard math and science, and assistants that carry out many steps on their own.
Gemini Embedding 2 turns text, images, video, audio and PDFs into fingerprints of their meaning in one shared space, up to about 8 pages per item. That lets one search find a picture from a sentence or a clip from a document. Pick it for search across kinds of media. For text-only collections of pieces under 2 pages, Gemini Embedding 001 is enough.
#8 for slide decks · Gemini 3 Flash Preview takes text, images, audio, video and PDFs, keeps about 1,000 pages in mind at once, and replies in text, reasoning before it answers. Google positions it for understanding mixed material, so give it screenshots, video clips and mixed documents to interpret, plus coding and assistant work that uses tools.
Gemini 3.1 Flash Lite takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. It is built for plain work at volume: translation, transcription, pulling out details, document processing and sending requests to the right model. It cannot operate a computer for you.
Gemma 4 31B is a model you can download and run yourself that uses all of itself on every word. It reads text and images and answers in text, keeps about 256 pages in mind at once, can think before it answers and uses your tools. Pick it when quality per answer matters more than speed.
Gemini Embedding 001 turns text into a fingerprint of its meaning, so a search finds text that means the same thing and not only the same words; it also helps sort text into categories. It takes text only, about 2 pages per piece. Pick it when your collection is plain text. For images, audio, video or PDFs in the same search, and for pieces up to about 8 pages, use Gemini Embedding 2.
Gemma 4 26B A4B is a model you can download and run yourself that uses only a small part of itself on each word, so it answers like a small model while aiming near the 31B in quality. It reads text and images, thinks on request, uses your tools, and keeps about 256 pages in mind at once. Pick it for fast work at high volume.
#6 for games · Agent work, coding and complex multi-step reasoning for jobs that need a quick answer that stays reliable over several steps, reading text, images, video, audio and files with tool calls and structured output
Lightweight reasoning built for ultra-low latency and high throughput, faster than earlier Flash models, with thinking off by default for speed and switchable on when a task needs more; the Lite of Gemini 2.5
Gemini 3.5 Flash Lite takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. It is the worker model for teams of assistants sharing a job and for reading documents apart: many small, focused tasks run fast. Use it under a stronger planner such as Gemini 3.8 Flash. Gemini 3.1 Flash Lite suits plain translation, and a full Flash model suits coding.
Reasoning, coding, maths and science from Google's 2.5 workhorse, with built-in thinking for more accurate answers and a cap on reasoning tokens you set per request; between 2.5 Flash-Lite and 2.5 Pro
Gemini 3.6 Flash takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. Google built it for writing code, assistants that work in steps, and web and app work, with fewer needless edits in what it writes. Pick it for everyday coding and assistant work. Gemini 3.8 Flash handles longer software-engineering tasks, and 3.5 Flash Lite handles high-volume simple jobs.
Google's preview of 3.1 Flash Lite takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. Google says it beats 2.5 Flash Lite and gets close to 2.5 Flash on overall quality, for large volumes of work such as pulling out details, translation and sorting requests.
#4 for slide decks · Gemini 3.1 Pro Preview takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and replies in text up to 65 pages long. Google tuned it for software engineering and for tasks it carries out on its own that need precise use of your tools and dependable multi-step execution. Pick it for hard coding jobs.
#2 for slide decks · A Gemini Flash that reaches close to Pro-level coding and reasoning, tuned for writing code and running agent loops in parallel, reading text, images, video, audio and PDFs, thinking at medium effort by default
Gemma 3 27B is the largest of the Gemma 3 models you can download and run yourself: it reads text and images, answers in text up to about 118 pages long, and uses your tools. It keeps about 128 pages in mind and speaks 140+ languages. Pick it for the hardest Gemma 3 jobs, such as reasoning, math and long answers.
Google's generally available 2.5 Pro, a thinking model that reasons through each answer for accuracy in hard reasoning, coding, maths and science, and took first place on the LMArena leaderboard
Gemma 3 12B is a model you can download and run yourself that reads text and images and answers in text, keeping about 128 pages in mind at once, in 140+ languages, and can use your tools. It sits in the middle of the Gemma 3 line, between the 4B and the 27B, and unlike the 4B it can use tools. Pick it for self-hosted assistants and image questions.
#5 for games · Gemini 3.8 Flash takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text up to 65 pages long, with low, medium and high thinking levels. Google rates it the strongest Flash model for long software-engineering runs, assistants that act on their own and business workflows.
Gemini 3.1 Flash Lite takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. It is built for plain work at volume: translation, transcription, pulling out details, document processing and sending requests to the right model. It cannot operate a computer for you.
Gemma 3 4B is the smallest Gemma 3 model you can download and run yourself: it reads text and images, answers in text, keeps about 128 pages in mind at once and speaks 140+ languages. Unlike the 12B and 27B, it cannot use tools. Pick it for light jobs such as short chat and simple image questions.
#1 for games · Gemini 4 Argon takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind, and answers in text up to about 1,000 pages long, with low, medium and high thinking. Google calls it its top model for reasoning. Pick it for complex software engineering, hard math and science, and assistants that carry out many steps on their own.
Gemini 3.5 Flash Lite takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. It is the worker model for teams of assistants sharing a job and for reading documents apart: many small, focused tasks run fast. Use it under a stronger planner such as Gemini 3.8 Flash. Gemini 3.1 Flash Lite suits plain translation, and a full Flash model suits coding.
The Custom Tools version of Google's 3.1 Pro tunes how it picks tools: it stops reaching for a generic command-line tool when one of your own, more efficient tools fits. It takes text, images, video, audio and PDFs and replies in text. Pick it for assistants that combine the command line with custom tools.
#6 for games · Agent work, coding and complex multi-step reasoning for jobs that need a quick answer that stays reliable over several steps, reading text, images, video, audio and files with tool calls and structured output
Gemini 3.6 Flash takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and returns text. Google built it for writing code, assistants that work in steps, and web and app work, with fewer needless edits in what it writes. Pick it for everyday coding and assistant work. Gemini 3.8 Flash handles longer software-engineering tasks, and 3.5 Flash Lite handles high-volume simple jobs.
Gemma 2 27B is a Google model you can download and run yourself that takes text and returns text, keeping about 8 pages in mind at once with replies up to 2 pages. That short memory and the lack of tool use make it a fit for self-contained chat, question answering and short summaries.
#4 for slide decks · Gemini 3.1 Pro Preview takes text, images, video, audio and PDFs, keeps about 1,000 pages in mind at once, and replies in text up to 65 pages long. Google tuned it for software engineering and for tasks it carries out on its own that need precise use of your tools and dependable multi-step execution. Pick it for hard coding jobs.
#8 for slide decks · Gemini 3 Flash Preview takes text, images, audio, video and PDFs, keeps about 1,000 pages in mind at once, and replies in text, reasoning before it answers. Google positions it for understanding mixed material, so give it screenshots, video clips and mixed documents to interpret, plus coding and assistant work that uses tools.
#2 for slide decks · A Gemini Flash that reaches close to Pro-level coding and reasoning, tuned for writing code and running agent loops in parallel, reading text, images, video, audio and PDFs, thinking at medium effort by default
This is the 5 June preview version of Gemini 2.5 Pro. It takes text, images, audio and files, keeps about 1,000 pages in mind, replies in text up to 65 pages long, and reasons through math, science and code before answering. Keep this one only to reproduce results you already have.