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.
Tetap teratur dengan koleksi
Simpan dan kategorikan konten berdasarkan preferensi Anda.
Saat berinteraksi dengan Gemini di Vertex AI, permintaan
(perintah) atau respons yang dihasilkan mungkin diblokir oleh filter konten. Dokumen
ini memberikan panduan tentang cara mengidentifikasi dan memproses respons
yang diblokir ini.
Kapan konten diblokir?
Ada dua skenario untuk filter konten yang memblokir konten:
Perintah input diblokir.
Respons model diblokir.
Bagian berikut menjelaskan cara memproses setiap skenario.
Perintah input diblokir
Jika perintah input Anda diblokir oleh filter konten sebelum dikirim ke
model, respons API akan menunjukkan karakteristik berikut:
Kolom
promptFeedback
diisi. Kolom ini berisi informasi tentang alasan
perintah diblokir.
[[["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,["# Process blocked responses\n\nWhen interacting with Gemini on Vertex AI, your requests\n(prompts) or the generated responses might be blocked by content filters. This\ndocument provides guidance on how to identify and process these blocked\nresponses.\n\nWhen is content blocked?\n------------------------\n\nThere are two scenarios for content filters blocking content:\n\n- The input prompt is blocked.\n- The model's response is blocked.\n\nThe sections that follow describe how to process each scenario.\n\n### The input prompt is blocked\n\nIf your input prompt is blocked by content filters before it is sent to the\nmodel, the API response exhibits the following characteristics:\n\n- The [`promptFeedback`](/vertex-ai/generative-ai/docs/reference/rest/v1/GenerateContentResponse#PromptFeedback) field is populated. This field contains information about why the prompt is blocked.\n- The [`candidates`](/vertex-ai/generative-ai/docs/reference/rest/v1/GenerateContentResponse) field isn't set.\n\n### The model's response is blocked\n\nIf the model generates a response but that response is blocked by content\nfilters, the API response exhibits the following characteristics:\n\n- The [`promptFeedback`](/vertex-ai/generative-ai/docs/reference/rest/v1/GenerateContentResponse#PromptFeedback) field isn't set.\n- The [`candidates`](/vertex-ai/generative-ai/docs/reference/rest/v1/GenerateContentResponse) field is set with the following characteristics:\n - The [`content`](/vertex-ai/generative-ai/docs/reference/rest/v1/Content) field isn't set.\n - The [`finishReason`](/vertex-ai/generative-ai/docs/reference/rest/v1/GenerateContentResponse#FinishReason) field is set and indicates why response generation stopped.\n\nYou can determine why the content was blocked by inspecting these fields in the\nAPI response.\n\nWhat's next\n-----------\n\n- Learn how to [configure content filters](/vertex-ai/generative-ai/docs/multimodal/configure-safety-filters)."]]
