diff --git a/openapi.yaml b/openapi.yaml
index a316a90fb..3fe227c53 100644
--- a/openapi.yaml
+++ b/openapi.yaml
@@ -22735,28 +22735,22 @@ components:
           x-linode-filterable: true
           type: string
           readOnly: true
-          description: >
-            The class of the Linode Type. We currently offer five classes of Linodes:
-
-              * nanode - Nanode instances are good for low-duty workloads,
-                where performance isn't critical. **Note:** As of June 16th, 2020, Nanodes became
-                1 GB Linodes in the Cloud Manager, however, the API, the CLI, and billing will
-                continue to refer to these instances as Nanodes.
-              * standard - Standard Shared instances are good for medium-duty workloads and
-                are a good mix of performance, resources, and price. **Note:** As of June 16th, 2020,
-                Standard Linodes in the Cloud Manager became Shared Linodes, however, the API, the CLI, and
-                billing will continue to refer to these instances as Standard Linodes.
-              * dedicated - Dedicated CPU instances are good for full-duty workloads
-                where consistent performance is important.
-              * gpu - Linodes with dedicated NVIDIA Quadro ® RTX 6000 GPUs accelerate highly
-                specialized applications such as machine learning, AI, and video transcoding.
-              * highmem - High Memory instances favor RAM over other resources, and can be
-                good for memory hungry use cases like caching and in-memory databases.
-                All High Memory plans contain dedicated CPU cores.
+          description: |
+            The class of the Linode Type.
+
+            We currently offer six classes of compute instances:
+
+              * `nanode` - Nanode instances are good for low-duty workloads, where performance isn't critical. **Note:** As of June 16th, 2020, Nanodes became 1 GB Linodes in the Cloud Manager, however, the API, the CLI, and billing will continue to refer to these instances as Nanodes.
+              * `standard` - Standard Shared instances are good for medium-duty workloads and are a good mix of performance, resources, and price. **Note:** As of June 16th, 2020, Standard Linodes in the Cloud Manager became Shared Linodes, however, the API, the CLI, and billing will continue to refer to these instances as Standard Linodes.
+              * `dedicated` - Dedicated CPU instances are good for full-duty workloads where consistent performance is important.
+              * `premium` (limited Regions) - In addition to the features of Dedicated instances, Premium instances come equipped with the latest AMD EPYC™ CPUs, ensuring your applications are running on the latest hardware with consistently high performance. Only available in [Regions](/docs/api/regions/#regions-list) with "Premium Plans" in their `capabilities`
+              * `gpu` (limited Regions) - Linodes with dedicated NVIDIA Quadro® RTX 6000 GPUs accelerate highly specialized applications such as machine learning, AI, and video transcoding. Only available in [Regions](/docs/api/regions/#regions-list) with "GPU Linodes" in their `capabilities`
+              * `highmem` - High Memory instances favor RAM over other resources, and can be good for memory hungry use cases like caching and in-memory databases. All High Memory plans contain dedicated CPU cores.
           enum:
           - nanode
           - standard
           - dedicated
+          - premium
           - gpu
           - highmem
           example: standard