ドキュメント
テキストからドキュメントを作成
生のテキストからナレッジベースにドキュメントを作成します。インデックスは非同期で実行されます。返された batch ID を使って ドキュメント埋め込みステータス（進捗）を取得 で追跡します。
POST
/
datasets
/
{dataset_id}
/
document
/
create-by-text
テキストからドキュメントを作成
curl --request POST \
--url https://{api_base_url}/datasets/{dataset_id}/document/create-by-text \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"indexing_technique": "high_quality",
"name": "guide.txt",
"text": "Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents."
}
'import requests
url = "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text"
payload = {
"indexing_technique": "high_quality",
"name": "guide.txt",
"text": "Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents."
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
indexing_technique: 'high_quality',
name: 'guide.txt',
text: 'Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.'
})
};
fetch('https://{api_base_url}/datasets/{dataset_id}/document/create-by-text', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'indexing_technique' => 'high_quality',
'name' => 'guide.txt',
'text' => 'Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.'
]),
CURLOPT_HTTPHEADER => [
"Authorization: Bearer <token>",
"Content-Type: application/json"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text"
payload := strings.NewReader("{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("Authorization", "Bearer <token>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://{api_base_url}/datasets/{dataset_id}/document/create-by-text")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://{api_base_url}/datasets/{dataset_id}/document/create-by-text")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}"
response = http.request(request)
puts response.read_body{
"batch": "20250306150245647595",
"document": {
"archived": false,
"created_at": 1741267200,
"created_by": "ad313dd6-ef04-4dd1-a5b0-c0f0b9e2e7e4",
"created_from": "api",
"data_source_detail_dict": {
"upload_file": {
"created_at": 1741267200,
"created_by": "ad313dd6-ef04-4dd1-a5b0-c0f0b9e2e7e4",
"extension": "txt",
"id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
"mime_type": "text/plain",
"name": "guide.txt",
"size": 2048
}
},
"data_source_info": {
"upload_file_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
},
"data_source_type": "upload_file",
"dataset_process_rule_id": "e1f2a3b4-c5d6-7890-ef12-345678901234",
"disabled_at": null,
"disabled_by": null,
"display_status": "indexing",
"doc_form": "text_model",
"doc_metadata": [],
"enabled": true,
"error": null,
"hit_count": 0,
"id": "a8e0e5b5-78c6-4130-a5ce-25feb0e0b4ac",
"indexing_status": "indexing",
"name": "guide.txt",
"need_summary": false,
"position": 1,
"summary_index_status": null,
"tokens": 0,
"word_count": 0
}
}Authorizations
すべてのリクエストは、Authorization: Bearer {API_KEY} ヘッダーに指定した API キーで認証します。アプリのエンドポイントにはアプリの API キーを、ナレッジのエンドポイントにはナレッジベースの API キーを使用します（Dify API クイックスタート）。
キーはサーバーサイドで保管し、クライアントコードには決して埋め込まないでください。キーがない、または無効な場合は HTTP 401（unauthorized）が返ります。
Path Parameters
ナレッジベース ID。ナレッジベースリストを取得 で取得します。スコープを限定したキーでは、Dify の URL から ID をコピー してください。
Body
application/json
ドキュメント名です。
ドキュメントのテキスト内容です。
チャンク構造：text_model は標準テキストチャンキング、hierarchical_model は親子チャンク、qa_model は質問・回答ペアの抽出です。
ナレッジベースの既存のチャンク構造と一致させる必要があります。省略した場合は既存の構造が使われ、ナレッジベースにまだチャンク構造がない場合は text_model になります。
Available options:
text_model, hierarchical_model, qa_model 処理最適化のためのドキュメント言語です。
埋め込みモデル名です。利用可能なモデルを取得 で model_type=text-embedding を指定した際の model フィールドの値を使用します。
埋め込みモデルプロバイダーです。利用可能なモデルを取得 で model_type=text-embedding を指定した際の provider フィールドの値を使用します。
ナレッジベースに最初のドキュメントを追加する際に必須です。以降のドキュメントでは省略するとナレッジベースのインデックス方式を継承します。high_quality は埋め込みモデルによる精密検索、economy はキーワードベースのインデックスを使用します。
Available options:
high_quality, economy バージョン管理用の元ドキュメント ID です。ナレッジベースのドキュメントリストを取得 から取得します。
チャンキングの処理ルールです。
Show child attributes
Show child attributes
このナレッジベースをクエリする際のチャンクの検索方法とランキング方法を制御します。
Show child attributes
Show child attributes
Was this page helpful?
⌘I
テキストからドキュメントを作成
curl --request POST \
--url https://{api_base_url}/datasets/{dataset_id}/document/create-by-text \
--header 'Authorization: Bearer <token>' \
--header 'Content-Type: application/json' \
--data '
{
"indexing_technique": "high_quality",
"name": "guide.txt",
"text": "Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents."
}
'import requests
url = "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text"
payload = {
"indexing_technique": "high_quality",
"name": "guide.txt",
"text": "Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents."
}
headers = {
"Authorization": "Bearer <token>",
"Content-Type": "application/json"
}
response = requests.post(url, json=payload, headers=headers)
print(response.text)const options = {
method: 'POST',
headers: {Authorization: 'Bearer <token>', 'Content-Type': 'application/json'},
body: JSON.stringify({
indexing_technique: 'high_quality',
name: 'guide.txt',
text: 'Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.'
})
};
fetch('https://{api_base_url}/datasets/{dataset_id}/document/create-by-text', options)
.then(res => res.json())
.then(res => console.log(res))
.catch(err => console.error(err));<?php
$curl = curl_init();
curl_setopt_array($curl, [
CURLOPT_URL => "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text",
CURLOPT_RETURNTRANSFER => true,
CURLOPT_ENCODING => "",
CURLOPT_MAXREDIRS => 10,
CURLOPT_TIMEOUT => 30,
CURLOPT_HTTP_VERSION => CURL_HTTP_VERSION_1_1,
CURLOPT_CUSTOMREQUEST => "POST",
CURLOPT_POSTFIELDS => json_encode([
'indexing_technique' => 'high_quality',
'name' => 'guide.txt',
'text' => 'Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.'
]),
CURLOPT_HTTPHEADER => [
"Authorization: Bearer <token>",
"Content-Type: application/json"
],
]);
$response = curl_exec($curl);
$err = curl_error($curl);
curl_close($curl);
if ($err) {
echo "cURL Error #:" . $err;
} else {
echo $response;
}package main
import (
"fmt"
"strings"
"net/http"
"io"
)
func main() {
url := "https://{api_base_url}/datasets/{dataset_id}/document/create-by-text"
payload := strings.NewReader("{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}")
req, _ := http.NewRequest("POST", url, payload)
req.Header.Add("Authorization", "Bearer <token>")
req.Header.Add("Content-Type", "application/json")
res, _ := http.DefaultClient.Do(req)
defer res.Body.Close()
body, _ := io.ReadAll(res.Body)
fmt.Println(string(body))
}HttpResponse<String> response = Unirest.post("https://{api_base_url}/datasets/{dataset_id}/document/create-by-text")
.header("Authorization", "Bearer <token>")
.header("Content-Type", "application/json")
.body("{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}")
.asString();require 'uri'
require 'net/http'
url = URI("https://{api_base_url}/datasets/{dataset_id}/document/create-by-text")
http = Net::HTTP.new(url.host, url.port)
http.use_ssl = true
request = Net::HTTP::Post.new(url)
request["Authorization"] = 'Bearer <token>'
request["Content-Type"] = 'application/json'
request.body = "{\n \"indexing_technique\": \"high_quality\",\n \"name\": \"guide.txt\",\n \"text\": \"Dify is an open-source LLM application development platform. This guide covers building AI applications with workflows, RAG pipelines, and agents.\"\n}"
response = http.request(request)
puts response.read_body{
"batch": "20250306150245647595",
"document": {
"archived": false,
"created_at": 1741267200,
"created_by": "ad313dd6-ef04-4dd1-a5b0-c0f0b9e2e7e4",
"created_from": "api",
"data_source_detail_dict": {
"upload_file": {
"created_at": 1741267200,
"created_by": "ad313dd6-ef04-4dd1-a5b0-c0f0b9e2e7e4",
"extension": "txt",
"id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890",
"mime_type": "text/plain",
"name": "guide.txt",
"size": 2048
}
},
"data_source_info": {
"upload_file_id": "a1b2c3d4-e5f6-7890-abcd-ef1234567890"
},
"data_source_type": "upload_file",
"dataset_process_rule_id": "e1f2a3b4-c5d6-7890-ef12-345678901234",
"disabled_at": null,
"disabled_by": null,
"display_status": "indexing",
"doc_form": "text_model",
"doc_metadata": [],
"enabled": true,
"error": null,
"hit_count": 0,
"id": "a8e0e5b5-78c6-4130-a5ce-25feb0e0b4ac",
"indexing_status": "indexing",
"name": "guide.txt",
"need_summary": false,
"position": 1,
"summary_index_status": null,
"tokens": 0,
"word_count": 0
}
}