diff --git a/.changeset/wild-owls-agree.md b/.changeset/wild-owls-agree.md
new file mode 100644
index 00000000..63547f84
--- /dev/null
+++ b/.changeset/wild-owls-agree.md
@@ -0,0 +1,5 @@
+---
+"@empiricalrun/ai": patch
+---
+
+fix: handle rate_limit_exceeded in openai assistants
diff --git a/packages/ai/src/providers/openai/index.ts b/packages/ai/src/providers/openai/index.ts
index 47e83911..4e0f3b0a 100644
--- a/packages/ai/src/providers/openai/index.ts
+++ b/packages/ai/src/providers/openai/index.ts
@@ -156,7 +156,11 @@ const runAssistant: ICreateAndRunAssistantThread = async (body) => {
           }
           return asstRunResp;
         })().catch((err: any) => {
-          if ((err.message as string).includes("server_error")) {
+          const serverError = (err.message as string).includes("server_error");
+          const rateLimitError = (err.message as string).includes(
+            "rate_limit_exceeded",
+          );
+          if (serverError || rateLimitError) {
             console.warn(
               `Retrying request due to server error (attempt ${attempt})`,
             );