Mulai 29 April 2025, model Gemini 1.5 Pro dan Gemini 1.5 Flash tidak tersedia di project yang belum pernah menggunakan model ini, termasuk project baru. Untuk mengetahui detailnya, lihat Versi dan siklus proses model.
Anda dapat menggunakan Vertex AI Studio untuk mendesain, menguji, dan mengelola perintah untuk model bahasa besar (LLM) Gemini Google dan model pihak ketiga. Vertex AI Studio mendukung model pihak ketiga tertentu yang ditawarkan di Vertex AI sebagai model sebagai layanan (MaaS), seperti model Claude dari Anthropic dan model Llama dari Meta.
Sebelum memulai
Panduan memulai ini mengharuskan Anda menyelesaikan langkah-langkah berikut untuk menyiapkan project Google Cloud dan mengaktifkan Vertex AI API.
Konsep
Perintah: Permintaan bahasa alami yang dikirimkan ke model bahasa untuk menghasilkan respons.
Vertex AI Studio: Alat konsol Google Cloud untuk mendesain, menguji, dan mengelola perintah untuk model bahasa besar.
Gemini: Serangkaian model bahasa besar multimodal yang dikembangkan oleh Google.
Prasyarat
Sign in to your Google Cloud account. If you're new to
Google Cloud,
create an account to evaluate how our products perform in
real-world scenarios. New customers also get $300 in free credits to
run, test, and deploy workloads.
In the Google Cloud console, on the project selector page,
select or create a Google Cloud project.
Perintah adalah permintaan bahasa alami yang dikirimkan ke model bahasa yang menghasilkan respons. Misalnya, perintah sederhana bisa berupa Summarize this article in three sentences: article-text.
Perintah dapat berisi pertanyaan, petunjuk, informasi kontekstual, contoh sedikit tembakan, dan input parsial agar model dapat diselesaikan. Setelah menerima perintah, bergantung pada jenis model yang digunakan, model dapat menghasilkan teks, penyematan, kode, gambar, video, musik, dan lainnya.
Contoh perintah di galeri perintah Vertex AI Studio telah didesain sebelumnya untuk membantu menunjukkan kemampuan model. Setiap perintah telah dikonfigurasi dengan model dan parameter value yang ditentukan, sehingga Anda dapat membuka contoh perintah dan mengklik Submit untuk menghasilkan respons.
Menguji model flash Gemini menggunakan perintah teks ringkasan
Mengirim perintah teks ringkasan ke Gemini API di Vertex AI. Tugas meringkas mengekstrak informasi paling penting dari teks. Anda dapat memberikan informasi dalam perintah untuk membantu model membuat ringkasan, atau meminta model untuk membuat ringkasannya sendiri.
1. Buka halaman Galeri perintah dari bagian Vertex AI di konsol Google Cloud . Buka galeri perintah
2. Di menu drop-down Tugas, pilih Ringkas.
3. Buka kartu Ringkasan audio. Contoh perintah ini menyertakan file audio dan meminta ringkasan konten file dalam daftar berbutir.
4. Perhatikan bahwa di panel setelan, nilai default model ditetapkan ke Gemini-2.0-flash-001. Anda dapat memilih model Gemini lain dengan mengklik Ganti model.
5. Klik Kirim untuk membuat ringkasan.
Output ditampilkan dalam respons.
6. Untuk melihat kode Vertex AI API yang digunakan untuk membuat ringkasan transkrip, klik Build with code > Get code.
Di panel Get code, Anda dapat memilih bahasa pilihan untuk mendapatkan kode contoh perintah, atau Anda dapat membuka kode Python di notebook Colab Enterprise.
Menguji model Gemini Flash menggunakan perintah pembuatan kode
Mengirim perintah pembuatan kode ke Gemini API di Vertex AI. Tugas pembuatan kode menghasilkan kode menggunakan deskripsi natural language.
1. Buka halaman Galeri perintah dari bagian Vertex AI di konsol Google Cloud . Buka galeri perintah
2. Di menu drop-down Tasks, pilih Code.
3. Buka kartu Buat kode dari komentar. Contoh perintah ini mencakup petunjuk sistem yang memberi tahu model cara merespons dan beberapa metode Java yang tidak lengkap.
4. Perhatikan bahwa di panel setelan, nilai default model ditetapkan ke Gemini-2.0-flash-001. Anda dapat memilih model Gemini lain dengan mengklik Ganti model.
5. Untuk menyelesaikan setiap metode dengan membuat kode di area yang ditandai <WRITE CODE HERE>, klik Submit.
Output ditampilkan dalam respons.
6. Untuk melihat kode Vertex AI API yang digunakan untuk membuat ringkasan transkrip, klik Build with code > Get code.
Di panel Get code, Anda dapat memilih bahasa pilihan untuk mendapatkan kode contoh perintah, atau Anda dapat membuka kode Python di notebook Colab Enterprise.
[[["Mudah dipahami","easyToUnderstand","thumb-up"],["Memecahkan masalah saya","solvedMyProblem","thumb-up"],["Lainnya","otherUp","thumb-up"]],[["Sulit dipahami","hardToUnderstand","thumb-down"],["Informasi atau kode contoh salah","incorrectInformationOrSampleCode","thumb-down"],["Informasi/contoh yang saya butuhkan tidak ada","missingTheInformationSamplesINeed","thumb-down"],["Masalah terjemahan","translationIssue","thumb-down"],["Lainnya","otherDown","thumb-down"]],["Terakhir diperbarui pada 2025-08-25 UTC."],[],[],null,["# Quickstart: Send text prompts to Gemini using Vertex AI Studio\n\nYou can use Vertex AI Studio to design, test, and manage prompts for\nGoogle's [Gemini](/vertex-ai/generative-ai/docs/overview) large language models\n(LLMs) and third-party models. Vertex AI Studio supports certain\nthird-party models that are offered on Vertex AI as [models as a\nservice (MaaS)](/vertex-ai/generative-ai/docs/partner-models/use-partner-models), such as\nAnthropic's Claude models and Meta's Llama models.\n| **Note:** On your initial use for third-party models, Vertex AI prompts you to accept the third-party's terms and conditions. You must do this once for each third-party provider to start using their models.\n\nIn this quickstart, you:\n\n- Send these prompts to the Gemini API using samples from the generative AI prompt gallery, including the following:\n - A summarization text prompt\n - A code generation prompt\n- View the code used to generate the responses\n\nBefore you begin prompting in Vertex AI Studio\n----------------------------------------------\n\nThis quickstart requires you to complete the following steps to set up a\nGoogle Cloud project and enable the Vertex AI API.\n\n- Sign in to your Google Cloud account. If you're new to Google Cloud, [create an account](https://console.cloud.google.com/freetrial) to evaluate how our products perform in real-world scenarios. New customers also get $300 in free credits to run, test, and deploy workloads.\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com)\n\n- In the Google Cloud console, on the project selector page,\n select or create a Google Cloud project.\n\n | **Note**: If you don't plan to keep the resources that you create in this procedure, create a project instead of selecting an existing project. After you finish these steps, you can delete the project, removing all resources associated with the project.\n\n [Go to project selector](https://console.cloud.google.com/projectselector2/home/dashboard)\n-\n [Verify that billing is enabled for your Google Cloud project](/billing/docs/how-to/verify-billing-enabled#confirm_billing_is_enabled_on_a_project).\n\n-\n\n\n Enable the Vertex AI API.\n\n\n [Enable the API](https://console.cloud.google.com/flows/enableapi?apiid=aiplatform.googleapis.com)\n\n\u003cbr /\u003e\n\nSample prompts in Vertex AI Studio\n----------------------------------\n\nA prompt is a natural language request submitted to a language model that\ngenerates a response. Prompts can contain questions, instructions, contextual\ninformation, [few-shot examples](/vertex-ai/generative-ai/docs/learn/prompts/introduction-prompt-design#few-shot-examples),\nand partial input for the model to complete. After the model receives a prompt,\ndepending on the type of model used, it can generate text, embeddings, code,\nimages, videos, music, and more.\n\nThe sample prompts in Vertex AI Studio [prompt gallery](/vertex-ai/generative-ai/docs/prompt-gallery)\nare predesigned to help demonstrate model capabilities. Each prompt is\npreconfigured with specified model and parameter values so you can open the\nsample prompt and click **Submit** to generate a response.\n\nTest the Gemini flash model using a summarization text prompt\n-------------------------------------------------------------\n\nSend a summarization text prompt to the Gemini API in Vertex AI. A summarization\ntask extracts the most important information from text. You can provide\ninformation in the prompt to help the model create a summary, or ask the model\nto create a summary on its own.\n\n1. Go to the **Prompt gallery** page from the Vertex AI\n section in the Google Cloud console. \n\n [Go to prompt gallery](https://console.cloud.google.com/vertex-ai/studio/prompt-gallery)\n\n2. In the **Tasks** drop-down menu, select **Summarize**.\n\n3. Open the **Audio summarization** card.\n\n This sample prompt includes an audio file and requests a summary of the file\n contents in a bulleted list.\n\n4. Notice that in the settings panel, the model's default value is set to\n **Gemini-2.0-flash-001** . You can choose a different Gemini model\n by clicking **Switch model**.\n\n5. Click **Submit** to generate the summary.\n\n The output is displayed in the response.\n6. To view the Vertex AI API code used to generate the transcript\n summary, click **Build with code** \\\u003e **Get code**.\n\n In the **Get code** panel, you can choose your preferred language to get the\n sample code for the prompt, or you can open the Python code in a\n Colab Enterprise notebook.\n\nTest the Gemini flash model using a code generation prompt\n----------------------------------------------------------\n\nSend a code generation prompt to the Gemini API in Vertex AI. A code generation task generates code\nusing a natural language description.\n\n1. Go to the **Prompt gallery** page from the Vertex AI\n section in the Google Cloud console. \n\n [Go to prompt gallery](https://console.cloud.google.com/vertex-ai/studio/prompt-gallery)\n\n2. In the **Tasks** drop-down menu, select **Code**.\n\n3. Open the **Generate code from comments** card.\n\n This sample prompt includes a [system instruction](/vertex-ai/generative-ai/docs/learn/prompts/system-instruction-introduction)\n that tells the model how to respond and some incomplete Java methods.\n\n4. Notice that in the settings panel, the model's default value is set to\n **Gemini-2.0-flash-001** . You can choose a different Gemini model\n by clicking **Switch model**.\n\n5. To complete each method by generating code in the areas marked\n `\u003cWRITE CODE HERE\u003e`, click **Submit** .\n\n The output is displayed in the response.\n6. To view the Vertex AI API code used to generate the transcript\n summary, click **Build with code** \\\u003e **Get code**.\n\n In the **Get code** panel, you can choose your preferred language to get the\n sample code for the prompt, or you can open the Python code in a\n Colab Enterprise notebook.\n\nDiscover what's next with prompts\n---------------------------------\n\n- See an [introduction to prompt design](/vertex-ai/generative-ai/docs/learn/prompts/introduction-prompt-design).\n- Learn about [designing multimodal prompts](/vertex-ai/generative-ai/docs/multimodal/design-multimodal-prompts) and [chat prompts](/vertex-ai/generative-ai/docs/chat/chat-prompts)."]]
