diff --git a/.github/workflows/opencode-review-dispatch.yml b/.github/workflows/opencode-review-dispatch.yml
index d2a0d1cc6..94f5cbb7c 100644
--- a/.github/workflows/opencode-review-dispatch.yml
+++ b/.github/workflows/opencode-review-dispatch.yml
@@ -3758,6 +3758,7 @@ jobs:
           OPENCODE_DYNAMIC_RUN_TIMEOUT_CAP_SECONDS: "5400"
           OPENCODE_DYNAMIC_TOTAL_BUDGET_CAP_SECONDS: "11700"
           OPENCODE_DYNAMIC_MAX_CYCLES_CAP: "0"
+          OPENCODE_FREE_RUN_TIMEOUT_SECONDS: "600"
           # This installation currently reports a 4k request-body limit for
           # GitHub Models GPT-5 endpoints even though the public catalog is
           # larger. Keep the exact runtime failure visible without spending a
diff --git a/scripts/ci/run_opencode_review_model_pool.sh b/scripts/ci/run_opencode_review_model_pool.sh
index b709ca807..3c4583069 100644
--- a/scripts/ci/run_opencode_review_model_pool.sh
+++ b/scripts/ci/run_opencode_review_model_pool.sh
@@ -381,6 +381,9 @@ cap_model_run_timeout() {
 	local cap_seconds
 
 	case "$model_candidate" in
+	opencode-free/*)
+		cap_seconds="$(env_integer_or_default OPENCODE_FREE_RUN_TIMEOUT_SECONDS 600)"
+		;;
 	github-models/openai/gpt-5 | github-models/openai/gpt-5-chat)
 		cap_seconds="$(env_integer_or_default OPENCODE_GITHUB_GPT5_RUN_TIMEOUT_SECONDS 45)"
 		;;
@@ -610,7 +613,7 @@ main() {
 				uncapped_run_timeout="$OPENCODE_RUN_TIMEOUT_SECONDS"
 				OPENCODE_RUN_TIMEOUT_SECONDS="$(cap_model_run_timeout "$model_candidate" "$OPENCODE_RUN_TIMEOUT_SECONDS")"
 				if [ "$OPENCODE_RUN_TIMEOUT_SECONDS" -lt "$uncapped_run_timeout" ]; then
-					printf 'OpenCode %s runtime cap selected %ss instead of %ss because this installation has returned a constrained request-body limit for that endpoint.\n' \
+					printf 'OpenCode %s runtime cap selected %ss instead of %ss because this provider has a bounded failover window.\n' \
 						"$model_candidate" "$OPENCODE_RUN_TIMEOUT_SECONDS" "$uncapped_run_timeout"
 				fi
 				export OPENCODE_RUN_TIMEOUT_SECONDS
diff --git a/tests/test_opencode_agent_contract.py b/tests/test_opencode_agent_contract.py
index fb9e6d3fb..0e791907f 100644
--- a/tests/test_opencode_agent_contract.py
+++ b/tests/test_opencode_agent_contract.py
@@ -1293,6 +1293,7 @@ def test_workflow_provisions_sandbox_tool_and_reviewer_agent():
     assert 'OPENCODE_DYNAMIC_RUN_TIMEOUT_CAP_SECONDS: "5400"' in workflow
     assert 'OPENCODE_DYNAMIC_TOTAL_BUDGET_CAP_SECONDS: "11700"' in workflow
     assert 'OPENCODE_DYNAMIC_MAX_CYCLES_CAP: "0"' in workflow
+    assert 'OPENCODE_FREE_RUN_TIMEOUT_SECONDS: "600"' in workflow
     assert 'OPENCODE_GITHUB_GPT5_RUN_TIMEOUT_SECONDS: "45"' in workflow
     assert 'OPENCODE_DYNAMIC_MAX_CYCLES: "0"' in workflow
     assert 'OPENCODE_BACKOFF_MAX_SECONDS: "30"' in workflow
@@ -1351,7 +1352,7 @@ def test_workflow_provisions_sandbox_tool_and_reviewer_agent():
     assert "while :" in model_pool_runner
     assert "should_skip_model_candidate" in model_pool_runner
     assert "cap_model_run_timeout" in model_pool_runner
-    assert "constrained request-body limit" in model_pool_runner
+    assert "bounded failover window" in model_pool_runner
     assert "run_central_adversarial_harness" not in model_pool_runner
     assert "finish_pool_without_model" in model_pool_runner
     assert "central-current-head-adversarial-harness" not in model_pool_runner
diff --git a/tests/test_opencode_model_pool_runner.py b/tests/test_opencode_model_pool_runner.py
index 5e8233487..3fd432242 100644
--- a/tests/test_opencode_model_pool_runner.py
+++ b/tests/test_opencode_model_pool_runner.py
@@ -759,7 +759,7 @@ def test_github_gpt5_runtime_cap_preserves_queue_budget(tmp_path: Path) -> None:
     assert result.returncode == 1
     assert (
         "OpenCode github-models/openai/gpt-5 runtime cap selected 3s instead of 9s "
-        "because this installation has returned a constrained request-body limit for that endpoint."
+        "because this provider has a bounded failover window."
     ) in result.stdout
     attempt_budget = re.search(
         r"OpenCode github-models/openai/gpt-5 attempt 1/1 using (\d+)s run timeout "
@@ -772,6 +772,24 @@ def test_github_gpt5_runtime_cap_preserves_queue_budget(tmp_path: Path) -> None:
     assert run_timeout <= remaining_budget <= 30
 
 
+def test_free_provider_runtime_cap_preserves_queue_budget(tmp_path: Path) -> None:
+    """A stalled free provider cannot consume a full paid-provider cadence slot."""
+    result = run_failed_model(
+        tmp_path,
+        extra_env={
+            "OPENCODE_FREE_RUN_TIMEOUT_SECONDS": "3",
+            "OPENCODE_RUN_TIMEOUT_SECONDS": "9",
+        },
+        model_candidates="opencode-free/nemotron-3-ultra-free",
+    )
+
+    assert result.returncode == 1
+    assert (
+        "OpenCode opencode-free/nemotron-3-ultra-free runtime cap selected 3s "
+        "instead of 9s because this provider has a bounded failover window."
+    ) in result.stdout
+
+
 def test_github_models_openai_prompt_references_evidence_without_inlining(
     tmp_path: Path,
 ) -> None: