FlowHunt
FlowHunt uses 16 subprocessors and infrastructure vendors, including AWS, Google Cloud and OpenAI, based on its public subprocessor list. Last checked Nov 14, 2026.
AI Stack
Web Search
Keyword, ranking, search, and backlink data for SEO and DataForSEO steps. Receives keywords, domains, and URLs.
Proxy network used to fetch YouTube transcripts and, when needed, web pages, screenshots, and videos. Requests, including the requested URLs, pass through its network.
Google search results for the search step. Receives search queries.
Model Serving
Routes requests to AI model providers. Used for open-weight and third-party models such as DeepSeek, Qwen, GLM (Z.ai), MiniMax, and MiMo, for the “FlowHunt Auto” options, and as a route to TypeSafe for classification. Receives prompts, instructions, conversation history, and context. The downstream recipient depends on OpenRouter’s routing for the model, and may be the model developer (for example DeepSeek, Alibaba, Z.ai, MiniMax, Google, OpenAI, or Anthropic) or another inference provider hosting the model (for example DeepInfra or Novita). For the “FlowHunt Auto” options, FlowHunt asks OpenRouter to use only inference providers that do not collect request data.
Amazon BedrockProcesses Anthropic Claude models, GPT-OSS models, and Amazon Nova models. Receives prompts, instructions, conversation history, and context, which may include knowledge source content, tool results, and data from connected platforms such as Google. Also processes the flow assistant that helps you build flows. This is separate from the hosting and storage described above.
Generates images, and in the PhotomaticAI app also videos, with models from developers such as Black Forest Labs, Google, ByteDance, Kling, Ideogram, Recraft, Luma, Runway, Stability AI, MiniMax, Alibaba, Tencent, and xAI. Receives image prompts and reference images, which can include image attachments in the session.
Foundation Models
Processes Claude models that are not served through Amazon Bedrock, and Claude models when Bedrock is unavailable or disabled for a model. Receives prompts, instructions, conversation history, and context.
Embeddings and data warehouse Google Vertex AI embeddings and Google BigQuery, as described above.
Processes Mistral models through the Mistral API (console.mistral.ai). Receives prompts, instructions, conversation history, and context.
Processes OpenAI models, including the default model of the AI agent step when no workspace default is set, and the fallback model of the classification step. Receives prompts, instructions, conversation history, and context. Also transcribes uploaded audio and video files (Whisper), reads scanned PDF pages as images when text recognition (OCR) is enabled, generates images with OpenAI image models (including session image attachments used as references), and receives campaign data you send through the OpenAI Ads integration.
Infrastructure
Cloud & Networking
AWSHosting, file storage, and backups Servers, file storage, and backups for the FlowHunt Service. AI model processing through Amazon Bedrock is described separately below.
Embeddings and data warehouse Google Vertex AI embeddings and Google BigQuery, as described above.
Billing & Payments
Changes
Checked Nov 14, 2026Source
This is the first saved copy of FlowHunt’s subprocessor list. Added and removed vendors will show here after the next check.
Similar companies
Companies with the most overlapping vendors