Google models

This guide provides an overview of the Google models available on Vertex AI. It covers the following model families:

  • Gemini models: Google's most capable multimodal models for complex reasoning, chat, and code generation.
  • Gemma models: A family of lightweight, state-of-the-art open models.
  • Embedding models: Models that convert text and multimodal data into vector representations.
  • Imagen models: Advanced models for image generation, editing, and captioning.
  • Veo models: Google's latest models for high-quality video generation.
  • MedLM models: Medically-tuned models for the healthcare industry.

Featured Gemini models

2.5 Pro

Our most advanced reasoning Gemini model, made to solve complex problems

  • Best for multimodal understanding
  • Capable of processing complex prompts and providing well-rounded responses
  • Best for coding, particularly for web development

2.5 Flash

Our best model in terms of price-performance, offering well-rounded capabilities

  • Support for Live API included for some endpoints
  • See the model's thinking process as part of the response
  • Balances price and performance

2.5 Flash-Lite

Our most cost effective model that supports high throughput tasks

  • The fastest model in the 2.5 line of models
  • Features a 1 million token context window and multimodal input, like 2.5 Flash
  • Outperforms 2.0 Flash on most evaluation benchmarks

All Gemini models

Model Description Status
Gemini 2.5 Pro Our most advanced reasoning model to date. GA
Gemini 2.5 Flash Our best model in terms of price-performance, offering well-rounded capabilities. GA
Gemini 2.5 Flash-Lite Our most cost effective model that supports high throughput tasks. Preview
Gemini 2.0 Flash Our newest multimodal model, with next generation features and improved capabilities. GA
Gemini 2.0 Flash-Lite A Gemini 2.0 Flash model optimized for cost efficiency and low latency. GA

Gemma models

Gemma is a family of lightweight, state-of-the-art open models built from the same research and technology used to create the Gemini models.

Model Description
Gemma 3 Our latest Gemma open model, featuring the ability to solve a wide variety of tasks with text and image input, support for over 140 languages, and long 128K context window.
Gemma 2 The second of generation of our open models featuring text generation, summarization, and extraction.
Gemma A small-sized, lightweight open model supporting text generation, summarization, and extraction.
ShieldGemma 2 Instruction tuned models for evaluating the safety of text and images against a set of defined safety policies.
PaliGemma Our open vision-language model that combines SigLIP and Gemma.
CodeGemma Powerful, lightweight open model that can perform a variety of coding tasks like fill-in-the-middle code completion, code generation, natural language understanding, mathematical reasoning, and instruction following.
TxGemma Generates predictions, classifications or text based on therapeutic related data and can be used to efficiently build AI models for therapeutic-related tasks with less data and less compute.

Embedding models

Model Description
Embeddings for Text Converts text data into vector representations for semantic search, classification, clustering, and similar tasks.
Multimodal Embeddings Generates vectors based on images, which can be used for downstream tasks like image classification, image search, and more.

Imagen models

Model Description Status
Imagen 3 for Generation Use text prompts to generate novel images. GA
Imagen 3 for Editing and Customization Use text prompts to edit existing input images, or parts of an image with a mask or generate new images based upon the context provided by input reference images. GA
Imagen 3 for Fast Generation Use text prompts to generate novel images with lower latency than our other image generation models. GA
Imagen for Captioning & VQA Use text prompts to generative novel images, edit existing ones, edit parts of an image with a mask and more. GA
Imagen 4 for Generation Use text prompts to generate novel images with higher quality than our previous image generation models. Preview
Imagen 4 for Fast Generation Use text prompts to generate novel images with higher quality and lower latency than our previous image generation models. Preview
Imagen 4 for Ultra Generation Use text prompts to generate novel images with higher quality and better prompt adherence than our previous image generation models. Preview

Veo models

Model Description Status
Veo 2 for Generation Use text prompts and images to generate novel videos. GA
Veo 3 for Generation Use text prompts and images to generate novel videos with higher quality than our previous video generation model. Preview
Veo 3 Fast for Generation Use text prompts and images to generate novel videos with higher quality and lower latency than our previous video generation model. Preview

MedLM models

Model Description
MedLM-medium HIPAA-compliant suite of medically tuned models designed to help healthcare practitioners with medical question and answer tasks, and summarization tasks for healthcare and medical documents.
MedLM-large-large HIPAA-compliant suite of medically tuned models designed to help healthcare practitioners with medical question and answer tasks, and summarization tasks for healthcare and medical documents.

Language support

Gemini

All the Gemini models can understand and respond in the following languages:

Afrikaans (af), Albanian (sq), Amharic (am), Arabic (ar), Armenian (hy), Assamese (as), Azerbaijani (az), Basque (eu), Belarusian (be), Bengali (bn), Bosnian (bs), Bulgarian (bg), Catalan (ca), Cebuano (ceb), Chinese (Simplified and Traditional) (zh), Corsican (co), Croatian (hr), Czech (cs), Danish (da), Dhivehi (dv), Dutch (nl), English (en), Esperanto (eo), Estonian (et), Filipino (Tagalog) (fil), Finnish (fi), French (fr), Frisian (fy), Galician (gl), Georgian (ka), German (de), Greek (el), Gujarati (gu), Haitian Creole (ht), Hausa (ha), Hawaiian (haw), Hebrew (iw), Hindi (hi), Hmong (hmn), Hungarian (hu), Icelandic (is), Igbo (ig), Indonesian (id), Irish (ga), Italian (it), Japanese (ja), Javanese (jv), Kannada (kn), Kazakh (kk), Khmer (km), Korean (ko), Krio (kri), Kurdish (ku), Kyrgyz (ky), Lao (lo), Latin (la), Latvian (lv), Lithuanian (lt), Luxembourgish (lb), Macedonian (mk), Malagasy (mg), Malay (ms), Malayalam (ml), Maltese (mt), Maori (mi), Marathi (mr), Meiteilon (Manipuri) (mni-Mtei), Mongolian (mn), Myanmar (Burmese) (my), Nepali (ne), Norwegian (no), Nyanja (Chichewa) (ny), Odia (Oriya) (or), Pashto (ps), Persian (fa), Polish (pl), Portuguese (pt), Punjabi (pa), Romanian (ro), Russian (ru), Samoan (sm), Scots Gaelic (gd), Serbian (sr), Sesotho (st), Shona (sn), Sindhi (sd), Sinhala (Sinhalese) (si), Slovak (sk), Slovenian (sl), Somali (so), Spanish (es), Sundanese (su), Swahili (sw), Swedish (sv), Tajik (tg), Tamil (ta), Telugu (te), Thai (th), Turkish (tr), Ukrainian (uk), Urdu (ur), Uyghur (ug), Uzbek (uz), Vietnamese (vi), Welsh (cy), Xhosa (xh), Yiddish (yi), Yoruba (yo), and Zulu (zu).

Gemma

Gemma supports only the English (en) language.

Embeddings

Multilingual text embedding models support the following languages:

Afrikaans (af), Albanian (sq), Amharic (am), Arabic (ar), Armenian (hy), Azerbaijani (az), Basque (eu), Belarusian (be), Bengali (bn), Bulgarian (bg), Catalan (ca), Cebuano (ceb), Chinese (Simplified and Traditional) (zh), Corsican (co), Czech (cs), Danish (da), Dutch (nl), English (en), Esperanto (eo), Estonian (et), Filipino (Tagalog) (fil), Finnish (fi), French (fr), Frisian (fy), Galician (gl), Georgian (ka), German (de), Greek (el), Gujarati (gu), Haitian Creole (ht), Hausa (ha), Hawaiian (haw), Hebrew (iw), Hindi (hi), Hmong (hmn), Hungarian (hu), Icelandic (is), Igbo (ig), Indonesian (id), Irish (ga), Italian (it), Japanese (ja), Javanese (jv), Kannada (kn), Kazakh (kk), Khmer (km), Korean (ko), Kurdish (ku), Kyrgyz (ky), Lao (lo), Latin (la), Latvian (lv), Lithuanian (lt), Luxembourgish (lb), Macedonian (mk), Malagasy (mg), Malay (ms), Malayalam (ml), Maltese (mt), Maori (mi), Marathi (mr), Mongolian (mn), Myanmar (Burmese) (my), Nepali (ne), Nyanja (Chichewa) (ny), Norwegian (no), Pashto (ps), Persian (fa), Polish (pl), Portuguese (pt), Punjabi (pa), Romanian (ro), Russian (ru), Samoan (sm), Scots Gaelic (gd), Serbian (sr), Sesotho (st), Shona (sn), Sindhi (sd), Sinhala (Sinhalese) (si), Slovak (sk), Slovenian (sl), Somali (so), Spanish (es), Sundanese (su), Swahili (sw), Swedish (sv), Tajik (tg), Tamil (ta), Telugu (te), Thai (th), Turkish (tr), Ukrainian (uk), Urdu (ur), Uzbek (uz), Vietnamese (vi), Welsh (cy), Xhosa (xh), Yiddish (yi), Yoruba (yo), and Zulu (zu).

Imagen 3

Imagen 3 supports the following languages:

English (en), Chinese (Simplified and Traditional) (zh), Hindi (hi), Japanese (ja), Korean (ko), Portuguese (pt), and Spanish (es).

MedLM

The MedLM model supports the English (en) language.

Explore all models in Model Garden

Model Garden is a platform that helps you discover, test, customize, and deploy Google proprietary and select OSS models and assets. To explore the generative AI models and APIs that are available on Vertex AI, go to Model Garden in the Google Cloud console.

Go to Model Garden

To learn more about Model Garden, including available models and capabilities, see Explore AI models in Model Garden.

Model versions

To see all model versions, including legacy and retired models, see Model versions and lifecycle.

What's next