IgnitionRAG vs Google Cloud (Vertex AI Search / RAG Engine)
Google Cloud offers Vertex AI Search (now Agent Search) and RAG Engine for building a RAG. IgnitionRAG provides the assistant ready to use, with cited sources, hosted in France on OVH.
Agent Search (formerly Vertex AI Search) creates search engines and generative answers over your data; RAG Engine orchestrates ingestion, embeddings and retrieval for your own applications. IgnitionRAG takes the ready-made platform route, with its connectors and interface.
Cost: an estimate
There are two routes. Agent Search bills per query (≈ €4 per 1,000 on Enterprise, basic generative answers included): about €18 per month for 5,000 questions and €89 to €134 for 25,000, not counting the application, its hosting and authentication. RAG Engine relies on a managed Spanner database in Paris (europe-west9): about €103 per month on the Basic tier and about €981 on the Scaled tier, Gemini 2.5 Flash model included. Estimates based on official prices on 14/10/2026; the query quota presented as a free trial is not taken into account. App hosting (Cloud Run) and authentication: not priced.
What you have to build yourself
A production document assistant is more than vector search. You need to create and synchronise data stores, manage access rights, display citations, develop or integrate the interface and authentication, isolate each client and wire up monitoring. Each component has to be built, tested and maintained. IgnitionRAG ships them together: collections, answers that cite their sources, permissions, interface and monitoring.
One point to check: data residency
The RAG Engine documentation states that data residency controls are not supported (VPC-SC and CMEK are). IgnitionRAG is hosted in France, on OVH.
When Google Cloud is the better choice
If your company is already on Google Cloud, with a team that knows the platform, building on Agent Search or RAG Engine is a consistent choice. IgnitionRAG's Agency/ESN licence deploys on the client's Azure, AWS or GCP, or with IgnitionAI, BYOK, with white-label, on quote. White-label is not included in Solo or Team.
At a glance
Official pages cloud.google.com (Agent Search pricing, RAG Engine billing), checked on 14/10/2026. Spanner Enterprise in Paris (europe-west9): $0.4756/replica-hour (SKU C02B-D5F3-E5D0, cloud.google.com, read on 14/10/2026) × 3 replicas = $1.4268/node-hour; Basic 100 PU: 0.1 × 1.4268 × 730 h ≈ $104; Scaled 1 node: 1.4268 × 730 h ≈ $1,042; + Gemini 2.5 Flash ≈ $11 / $55 → route B totals ≈ $115 / $1,097. Price list: cloud.google.com. Converted at the ECB reference rate of 13/10/2026: €1 = $1.1188 (ecb.europa.eu). $4 → ≈ €4; $20 → ≈ €18; $100 → ≈ €89; $150 → ≈ €134; $115 → ≈ €103; $1,097 → ≈ €981. App hosting (Cloud Run) and authentication: not priced. Estimate under stated assumptions: low = 10 users, 2,000 documents, 5,000 questions/month; realistic = 50 users, 20,000 documents, 25,000 questions/month; ~4,000 input tokens and ~400 output tokens per question. Development and maintenance not included.
| IgnitionRAG | Google Cloud (Vertex AI Search / RAG Engine) (to build) | |
|---|---|---|
| Getting started | Ready-made platform: collections, chat, citations, connectors | Components to assemble and maintain |
| Price | Solo €59/month excl. VAT · Team €249/month excl. VAT, AI credits included | Production estimate: ≈ €18 to €103/month (10 users) to ≈ €89 to €981/month (50 users) depending on the route, excluding application and development |
| Model API key | None on SaaS (Solo, Team): AI credits included, no key to provide; BYOK with the Agency/ESN licence | Your keys and your model billing |
| Connected sources | Web (scraping), Azure (SharePoint, Tables, Blob), AWS S3 + MinIO, GitHub, Hugging Face; WhatsApp (Unipile, beta) | Data stores (Cloud Storage, BigQuery, website…) |
| Source citations | Included | To implement and display |
| Permissions and multi-organisation | Built in | IAM / access control and per-client data store to design |
| Chat interface and authentication | Included | Agent Search widget or app to develop |
| Hosting | France, OVH (SaaS); Agency/ESN licence: client's Azure, AWS or GCP, or with IgnitionAI | Your Google Cloud project; data residency not supported by RAG Engine (official docs) |
| Agencies / ESN | Agency/ESN licence: deployment on the client's Azure, AWS or GCP, or with IgnitionAI, BYOK, white-label, on quote. Not in Solo or Team. | To build per client |
| Trial | 14-day free trial | Trial query quota (Agent Search) |
FAQ
Is it cheaper to build it yourself?
In production, no: under our assumptions, a Google Cloud deployment comes to about €18 to €103 per month for a small team and €89 to €981 for 50 users, before development and maintenance. IgnitionRAG: Solo €59/month excl. VAT, Team €249/month excl. VAT, AI credits included.
When is it better to build it yourself?
If you are already on Google Cloud, with a team available to build and maintain the application.
Do I need a model API key with IgnitionRAG?
No for Solo and Team: AI credits are included. The Agency/ESN licence works with BYOK, with white-label, deployed on the client's Azure, AWS or GCP or with IgnitionAI, on quote. White-label is not included in Solo or Team.
Does IgnitionRAG work with WhatsApp?
Yes, via Unipile, in beta.
How can I try it?
Join early access: 14-day free trial.
Deploy AI agents that work for your company.
Connect authorized knowledge and tools, then put an agent to work where your team needs it.