diff --git a/docs/sources/flow/requirements/_index.md b/docs/sources/flow/requirements/_index.md
new file mode 100644
index 000000000000..ad3b27c5806d
--- /dev/null
+++ b/docs/sources/flow/requirements/_index.md
@@ -0,0 +1,69 @@
+---
+aliases:
+- /docs/grafana-cloud/agent/flow/requirements
+- /docs/grafana-cloud/send-data/agent/flow/requirements
+canonical: https://grafana.com/docs/agent/latest/flow/requirements/
+description: Resource requirements
+menuTitle: Sizing
+title: Resource requirements
+weight: 100
+---
+
+# Resource requirements
+
+Capacity planning for {{< param "PRODUCT_NAME" >}} is a very important step in the deployment process. The sizing and capacity requirements for {{< param "PRODUCT_NAME" >}} depends on your specific environment, and the workload you have. Understanding the minimum hardware requirements for Grafana Agent Flow XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXx
+
+## Baseline
+
+The following values define the baseline for estimating the resource requirements for {{< param "PRODUCT_NAME" >}}. These values provide a starting point for your deployment requirements.
+
+| Data point | Size    | Throughput |
+|------------|---------|------------|
+| Metrics    | XXXXXXX | XXXXXXX    |
+| Events     | XXXXXXX | XXXXXXX    |
+| Logs       | XXXXXXX | XXXXXXX    |
+| Traces     | XXXXXXX | XXXXXXX    |
+
+A metrics series is active if it continuously has new data points appended to it. A metric series become inactive if they haven't received any new samples since the last WAL truncation interval.
+
+## Test environment
+
+Info about Agent version, software/OS, hardware etc. Definition of what a vCPU is?
+
+Does it make any difference running on Linux, Windows, OSX, etc.?
+
+### Test results for 14,000,000 to 16,000,000 metric series
+
+| vCPU cores | Memory         | Disk I/O                                        | Disk space       | Network |
+|------------|----------------|-------------------------------------------------|------------------| ------- |
+| 1.5 to 2.2 | 81 GB to 88 GB | 0.7%, with periodic spikes to 6%-8% for the WAL | 2.7 GB to 7.4 GB | XXXXXX  |
+
+### Test results for 650,000 to 800,000 metric series
+
+| vCPU cores | Memory           | Disk I/O                                       | Disk space       | Network |
+|------------|------------------|------------------------------------------------|------------------| ------- |
+| 0.1 to 0.2 | 3.9 GB to 5.7 GB | 0.3%, with periodic spikes to 0.4% for the WAL | 230 MB to 410 MB | XXXXXX  |
+
+## Recommendations
+
+Intro covering general recommendations.
+
+### CPUs
+
+Info about vCPUs for common use cases. For example X vCPUs when scraping some measurable number of metrics and sending over remote_write.
+
+### Memory
+
+Info about memory/RAM for common use cases. For example X GB RAM when scraping some measurable number of metrics and sending over remote_write.
+
+### Disk
+
+Info about disk space for common use cases. For example X GB of disk space when scraping some measurable number of metrics and sending over remote_write.
+
+### Network
+
+Info about network bandwidth for common use cases. For example X MBps when scraping some measurable number of metrics and sending over remote_write.
+
+## Clustering requirements
+
+Text