apunkt/dograh
1
0
Fork 0
mirror of https://github.com/dograh-hq/dograh.git synced 2026-07-10 11:12:13 +02:00
Code Issues Projects Releases Packages Wiki Activity

docs: add model provider data usage disclaimers

Browse source
This commit is contained in:
Sabiha Khan 2026-07-09 10:51:39 +05:30
parent c6c3f93d72
commit f3bcf24370
5 changed files with 38 additions and 1 deletions
Download patch file Download diff file Expand all files Collapse all files

6
docs/configurations/inference-providers.mdx
View file

@ -64,6 +64,12 @@ For details on creating and using Service Keys, see [API Keys and Service Keys](
Use **BYOK** when you want to bring your own provider accounts and API keys. This gives you separate control over each model category.
<Warning>
When you use BYOK or external model providers, Dograh sends only the data required for the selected service to operate. Depending on the provider and service type, this may include prompts, conversation history, transcripts, audio, generated text, tool/function definitions, tool inputs or results, and request metadata.
Provider data handling varies. Review each provider's data processing, retention, model training, and regional hosting policies before using sensitive data.
</Warning>
![BYOK model configuration](../images/model-configuration-byok.png)
The BYOK section has nested tabs:

4
docs/configurations/llm.mdx
View file

@ -5,6 +5,10 @@ description: "Voice Agents use LLM (Large Language Models), which are trained to
Dograh platform supports OpenAI, Google AI Studio, Google Vertex AI, Azure OpenAI, AWS Bedrock, Groq, OpenRouter, Hugging Face, MiniMax, Sarvam, and Dograh-managed LLMs. There are some models provided by default for you to choose from the drop down.
<Warning>
LLM providers receive the data needed to generate responses, such as prompts, conversation history, model settings, and any configured tool/function definitions or tool call context. Review the provider's data processing, retention, model training, and regional hosting policies before using sensitive data.
</Warning>
For locally deployed or self-hosted LLMs, Dograh also supports OpenAI-compatible endpoints such as Ollama and vLLM.
![Select Models from DropDown](../images/models_dropdown.png)

4
docs/configurations/transcriber.mdx
View file

@ -5,6 +5,10 @@ description: "Voice Agents use STT (Speech to Text), to transcribe what the user
Dograh platform supports Deepgram, OpenAI, Google, Azure Speech, AssemblyAI, Speechmatics, Cartesia, Gladia, Sarvam, Smallest AI, Hugging Face, and Dograh transcribers. You can take a look at the providers documentation of which language to select for your language requirements.
<Warning>
Transcriber providers receive the data needed to transcribe speech, such as call audio, language settings, keyterms, and request metadata. Review the provider's data processing, retention, model training, and regional hosting policies before using sensitive data.
</Warning>
For locally deployed or self-hosted STT models, Dograh also supports Speaches, an OpenAI API-compatible server for streaming transcription.
Example: Deepgram has their language support documentation at https://developers.deepgram.com/docs/models-languages-overview#nova-3

4
docs/configurations/voice.mdx
View file

@ -5,6 +5,10 @@ description: "Voice Agents use TTS (Text to Speech), which generates audio that
Dograh platform supports ElevenLabs, OpenAI, Google, Azure Speech, Deepgram, Cartesia, Smallest AI, MiniMax, Sarvam, Rime, Inworld, Camb.ai, xAI, and Dograh TTS engines. There are some voices from the providers that we ship by default. You can refer to the providers API documentation to select a voice ID that's most relevant for your language requirement.
<Warning>
Voice providers receive the data needed to synthesize speech, such as generated text, selected voice, model settings, and request metadata. Review the provider's data processing, retention, model training, and regional hosting policies before using sensitive data.
</Warning>
For locally deployed or self-hosted TTS models, Dograh also supports Speaches, an OpenAI API-compatible server for speech generation.
If you don't find your favourite voice, you can always add the voice ID manually.

21
ui/src/components/AIModelConfigurationV2Editor.tsx
View file

@ -1,6 +1,6 @@
"use client";
import { KeyRound, Save } from "lucide-react";
import { Info, KeyRound, Save } from "lucide-react";
import { useEffect, useMemo, useState } from "react";
import type { OrganizationAiModelConfigurationV2 } from "@/client/types.gen";
@ -265,6 +265,23 @@ function optionalByokService(config: Record<string, unknown>, service: ServiceSe
return serviceConfiguration;
}
function ThirdPartyProviderNotice() {
return (
<div className="mb-4 flex gap-3 rounded-md border border-amber-500/40 bg-amber-500/10 px-4 py-3 text-sm text-amber-900 dark:text-amber-200">
<Info className="mt-0.5 h-4 w-4 shrink-0" />
<div>
<p className="font-medium">Third-party provider data notice</p>
<p className="mt-1 leading-6">
Dograh sends data required by the selected model service. This may include prompts,
transcripts, audio, generated text, tool data, and request metadata depending on the
provider and service type. Review the provider&apos;s data and retention policies before
using sensitive data.
</p>
</div>
</div>
);
}
export function AIModelConfigurationV2Editor({
defaults,
configuration,
@ -383,6 +400,7 @@ export function AIModelConfigurationV2Editor({
<p className="mb-4 text-sm text-muted-foreground">
A single speech-to-speech model handles the conversation in realtime (no separate transcriber or voice). An LLM is still required for variable extraction and QA.
</p>
<ThirdPartyProviderNotice />
<ServiceConfigurationForm
key={`realtime-${JSON.stringify(realtimeInitialConfig)}`}
mode="global"
@ -472,6 +490,7 @@ export function AIModelConfigurationV2Editor({
</TabsContent>
<TabsContent value="byok" className="mt-0">
<ThirdPartyProviderNotice />
<ServiceConfigurationForm
key={`byok-${JSON.stringify(pipelineInitialConfig)}`}
mode="global"