What AI models does the agent use?

The agent has access to Claude, DeepSeek, Gemini, and other top models. It automatically routes to the best model for each task — you never have to think about API keys or token limits.

Yes. Each agent runs in an isolated environment. We don't train on your data, and you can export or delete everything at any time. The underlying engine (GoGogot) is open-source, so you can audit exactly how it works.

What can the agent actually do?

Browse the web, process files (CSV, PDF, images), run scheduled tasks, remember context across sessions, and communicate via Telegram or Slack. Think of it as a capable junior employee, not a chatbot.

How does memory work?

The agent maintains persistent memory across sessions. It remembers your preferences, past conversations, and task context. You can also explicitly tell it to remember or forget things.

Can I self-host the agent?

Absolutely. GoGogot is 100% open-source. Run it on your own hardware for free. cowork.ink is the managed version — we handle servers, model costs, updates, and uptime so you don't have to.

Is there a free trial?

We offer a 7-day money-back guarantee. Spin up an agent, give it real tasks, and if it doesn't save you time, we'll refund you — no questions asked.

Agentic RAG embeds an autonomous AI agent into the retrieval-augmented generation pipeline. Instead of a single retrieve-then-generate pass, the agent plans the retrieval strategy, issues multiple targeted queries, grades the results, and loops back to search again if the context is insufficient — before finally generating the answer.

How is Agentic RAG different from traditional RAG?

Traditional (naive) RAG converts the user query into a vector, retrieves the top-K chunks, and feeds them to the LLM once. Agentic RAG treats retrieval as a multi-step reasoning problem: the agent can decompose the query, choose from multiple retrieval tools, evaluate relevance, and iterate — resulting in significantly higher accuracy for complex or multi-hop questions.

When should I use Agentic RAG instead of traditional RAG?

Use Agentic RAG when queries are multi-step, require synthesizing information across multiple data sources, or demand high accuracy (legal, compliance, medical). Use traditional RAG for simple, single-source lookups where latency and cost must be minimised.

What frameworks support Agentic RAG?

LangGraph (stateful multi-step workflows), LangChain (agent + retriever tools), n8n (no-code agentic workflows), AWS Bedrock Agents, Vertex AI Agent Builder, and Weaviate all have first-class support for agentic RAG patterns. See our [guide to AI agent architecture](/blog/ai-agent-architecture/) for how these fit together.

What are the main trade-offs of Agentic RAG?

Agentic RAG delivers higher accuracy and can handle complex queries, but it introduces more latency (multiple retrieval rounds), higher token costs (each loop burns tokens), and more engineering complexity. For simple lookups, the overhead is not worth it — start with traditional RAG and add agentic loops only where accuracy demands it.

Agentic RAG: How AI Agents Supercharge Retrieval-Augmented Generation

Agentic RAG replaces static pipelines with autonomous agents that plan, retrieve, and self-correct. Guide to patterns, architecture & trade-offs.

Frequently Asked Questions

What is Agentic RAG?: Agentic RAG embeds an autonomous AI agent into the retrieval-augmented generation pipeline. Instead of a single retrieve-then-generate pass, the agent plans the retrieval strategy, issues multiple targeted queries, grades the results, and loops back to search again if the context is insufficient — before finally generating the answer.
How is Agentic RAG different from traditional RAG?: Traditional (naive) RAG converts the user query into a vector, retrieves the top-K chunks, and feeds them to the LLM once. Agentic RAG treats retrieval as a multi-step reasoning problem: the agent can decompose the query, choose from multiple retrieval tools, evaluate relevance, and iterate — resulting in significantly higher accuracy for complex or multi-hop questions.
When should I use Agentic RAG instead of traditional RAG?: Use Agentic RAG when queries are multi-step, require synthesizing information across multiple data sources, or demand high accuracy (legal, compliance, medical). Use traditional RAG for simple, single-source lookups where latency and cost must be minimised.
What frameworks support Agentic RAG?: LangGraph (stateful multi-step workflows), LangChain (agent + retriever tools), n8n (no-code agentic workflows), AWS Bedrock Agents, Vertex AI Agent Builder, and Weaviate all have first-class support for agentic RAG patterns. See our [guide to AI agent architecture](/blog/ai-agent-architecture/) for how these fit together.
What are the main trade-offs of Agentic RAG?: Agentic RAG delivers higher accuracy and can handle complex queries, but it introduces more latency (multiple retrieval rounds), higher token costs (each loop burns tokens), and more engineering complexity. For simple lookups, the overhead is not worth it — start with traditional RAG and add agentic loops only where accuracy demands it.