diff --git a/pages/generative-apis/menu.ts b/pages/generative-apis/menu.ts
index 2edb20b386..f20c875a76 100644
--- a/pages/generative-apis/menu.ts
+++ b/pages/generative-apis/menu.ts
@@ -147,6 +147,10 @@ export const generativeApisMenu = {
           label: 'Security and reliability in Generative APIs',
           slug: 'security-and-reliability',
         },
+        {
+          label: 'Understanding Generative APIs costs',
+          slug: 'cost-estimator',
+        },
         {
           label: 'Adding AI to VS Code using Continue',
           slug: 'adding-ai-to-vscode-using-continue',
diff --git a/pages/generative-apis/reference-content/cost-estimator.mdx b/pages/generative-apis/reference-content/cost-estimator.mdx
new file mode 100644
index 0000000000..b72af3d49b
--- /dev/null
+++ b/pages/generative-apis/reference-content/cost-estimator.mdx
@@ -0,0 +1,35 @@
+---
+title: Understanding Generative APIs costs
+description: Compare costs for the different deployment options of Scaleway Generative APIs - Serverless versus Dedicated Deployment.
+tags: generative-apis how-much cost charge
+dates:
+  validation: 2026-05-11
+  posted: 2026-05-11
+---
+
+Understanding the financial impact of AI workloads is essential for making informed decisions. The Cost estimator (available via the [Scaleway console](https://console.scaleway.com/generative-api/cost-estimator)) provides a clear, side‑by‑side view of how the same generative workload behaves under the different deployment options offered by Scaleway: [Generative APIs - Serverless versus Generative APIs Dedicated Deployment](/generative-apis/faq/#what-is-the-difference-between-serverless-and-dedicated-deployment).
+
+<Message type="note">
+  - The Cost estimator provides an estimate based on standard benchmarks, assuming significant concurrency and a low cache hit rate. Only performance tests in production, based on your actual workload, can provide a fully accurate estimate.
+  - Performance may vary significantly for extreme input/output ratios (e.g., 100:1 or 1:10). In these cases, processing is bottlenecked by either input-heavy or output-heavy workloads.
+  - For dedicated deployments, caching is implicit and exclusive to each user. This can significantly improve performance for use cases with many similar input tokens, such as a long system prompt with a common prefix shared across requests, typical of extended conversations.
+</Message>
+
+## Compare costs
+
+1. Log in to the [Scaleway console](https://console.scaleway.com/).
+
+2. Click **Generative APIs** in the **AI** section of the side menu.
+
+3. Select the **Cost estimator** tab.
+
+4. Model your workload by setting the following: 
+    - **Number of users**
+    - **Queries per user per day**
+    - **Hours of usage/day**
+    - **Load**
+
+5. Set your chosen **Model** and **GPU**.
+    The estimator instantly calculates the total monthly cost for both Serverless and Dedicated modes.
+
+6. Compare cost differences side‑by‑side. 
\ No newline at end of file