diff --git a/src/dstack/_internal/server/routers/runs.py b/src/dstack/_internal/server/routers/runs.py
index a6523a1ba..ab468c16d 100644
--- a/src/dstack/_internal/server/routers/runs.py
+++ b/src/dstack/_internal/server/routers/runs.py
@@ -10,6 +10,7 @@
 from dstack._internal.server.db import get_session
 from dstack._internal.server.models import ProjectModel, UserModel
 from dstack._internal.server.schemas.runs import (
+    MAX_JOB_SUBMISSIONS_LIMIT,
     ApplyRunPlanRequest,
     DeleteRunsRequest,
     GetRunPlanRequest,
@@ -61,10 +62,14 @@ async def list_runs(
     `project_name`, `repo_id`, `username`, and `only_active` can be specified as filters.
     Setting `only_active` to `true` excludes finished runs and deleted runs.
     Specifying `repo_id` without `project_name` returns no runs.
+    At most `job_submissions_limit` latest job submissions are returned per job.
 
     The results are paginated. To get the next page, pass `submitted_at` and `id` of
     the last run from the previous page as `prev_submitted_at` and `prev_run_id`.
     """
+    job_submissions_limit = body.job_submissions_limit
+    if job_submissions_limit is None:
+        job_submissions_limit = MAX_JOB_SUBMISSIONS_LIMIT
     run_list = await runs.list_user_runs(
         session=session,
         user=user,
@@ -73,7 +78,7 @@ async def list_runs(
         username=body.username,
         only_active=body.only_active,
         include_jobs=body.include_jobs,
-        job_submissions_limit=body.job_submissions_limit,
+        job_submissions_limit=job_submissions_limit,
         prev_submitted_at=body.prev_submitted_at,
         prev_run_id=body.prev_run_id,
         limit=body.limit,
diff --git a/src/dstack/_internal/server/schemas/runs.py b/src/dstack/_internal/server/schemas/runs.py
index 844724371..12b0c36ba 100644
--- a/src/dstack/_internal/server/schemas/runs.py
+++ b/src/dstack/_internal/server/schemas/runs.py
@@ -7,6 +7,8 @@
 from dstack._internal.core.models.common import CoreModel
 from dstack._internal.core.models.runs import ApplyRunPlanInput, RunSpec
 
+MAX_JOB_SUBMISSIONS_LIMIT = 10
+
 
 class ListRunsRequest(CoreModel):
     project_name: Optional[str] = None
@@ -20,9 +22,11 @@ class ListRunsRequest(CoreModel):
     job_submissions_limit: Optional[int] = Field(
         None,
         ge=0,
+        le=MAX_JOB_SUBMISSIONS_LIMIT,
         description=(
             "Limit number of job submissions returned per job to avoid large responses."
-            "Drops older job submissions. No effect with `include_jobs: false`"
+            " Drops older job submissions. No effect with `include_jobs: false`."
+            f" Defaults to the maximum allowed value of {MAX_JOB_SUBMISSIONS_LIMIT}"
         ),
     )
     prev_submitted_at: Optional[datetime] = None
diff --git a/src/tests/_internal/server/routers/test_runs.py b/src/tests/_internal/server/routers/test_runs.py
index e9aba1e8f..e98e0b150 100644
--- a/src/tests/_internal/server/routers/test_runs.py
+++ b/src/tests/_internal/server/routers/test_runs.py
@@ -48,7 +48,7 @@
 from dstack._internal.core.models.users import GlobalRole, ProjectRole
 from dstack._internal.core.models.volumes import InstanceMountPoint, MountPoint
 from dstack._internal.server.models import JobModel, RunModel
-from dstack._internal.server.schemas.runs import ApplyRunPlanRequest
+from dstack._internal.server.schemas.runs import MAX_JOB_SUBMISSIONS_LIMIT, ApplyRunPlanRequest
 from dstack._internal.server.services.projects import add_project_member
 from dstack._internal.server.services.resources import (
     set_gpu_vendor_default,
@@ -983,6 +983,47 @@ async def test_limits_job_submissions(
             },
         ]
 
+    @pytest.mark.asyncio
+    @pytest.mark.parametrize("test_db", ["sqlite", "postgres"], indirect=True)
+    async def test_limits_job_submissions_by_default(
+        self, test_db, session: AsyncSession, client: AsyncClient
+    ):
+        user = await create_user(session=session, global_role=GlobalRole.USER)
+        project = await create_project(session=session, owner=user)
+        await add_project_member(
+            session=session, project=project, user=user, project_role=ProjectRole.USER
+        )
+        repo = await create_repo(
+            session=session,
+            project_id=project.id,
+        )
+        run = await create_run(
+            session=session,
+            project=project,
+            repo=repo,
+            user=user,
+        )
+        submissions_num = MAX_JOB_SUBMISSIONS_LIMIT + 2
+        for submission_num in range(submissions_num):
+            await create_job(
+                session=session,
+                run=run,
+                submission_num=submission_num,
+            )
+        response = await client.post(
+            "/api/runs/list",
+            headers=get_auth_headers(user.token),
+            json={},
+        )
+        assert response.status_code == 200, response.json()
+        runs = response.json()
+        assert len(runs) == 1
+        job_submissions = runs[0]["jobs"][0]["job_submissions"]
+        assert len(job_submissions) == MAX_JOB_SUBMISSIONS_LIMIT
+        assert [js["submission_num"] for js in job_submissions] == list(
+            range(submissions_num - MAX_JOB_SUBMISSIONS_LIMIT, submissions_num)
+        )
+
     @pytest.mark.asyncio
     @pytest.mark.parametrize(
         "client_version,expected_probes",