fix(console): include legacy run input tokens in model usage (#7411)

README.md

@@ -254,6 +254,8 @@ fences, including unclosed fences, in complete and streamed replies.
    Optional per-model pricing must use one currency across all priced models.
    DeerFlow disables Console cost estimates when currencies are mixed rather
    than presenting an invalid aggregate.
+   For older runs without a per-model usage breakdown, Console attributes the
+   recorded total and input token counts to the run's model.
 
    Administrators can open **Settings → Models** to select a provider and enter
    only its **API Key** and **model ID**. Endpoints, authentication and adapters are

backend/app/gateway/routers/AGENTS.md

@@ -1,5 +1,10 @@
 # Router contracts
 
+Console usage attributes legacy rows without a per-model usage map to
+`model_name`, including their recorded `total_input_tokens`. Accumulate those
+inputs with newer usage buckets for the same model; do not infer cache hits.
+Regression coverage lives in `tests/test_console_router.py`.
+
 Shared scheduled-task update validation preserves an unchanged future `once`
 instant on a live task inside `scheduler.min_once_delay_seconds`; new/changed
 instants and terminal reactivation still enforce the minimum delay.

backend/app/gateway/routers/console.py

@@ -528,6 +528,7 @@ async def console_usage(
             # Legacy rows predating token_usage_by_model: fall back to the run's model.
             entry = by_model.setdefault(row.model_name, ConsoleUsageModelBreakdown())
             entry.tokens += run_tokens
+            entry.input_tokens += row.total_input_tokens or 0
             entry.runs += 1
             if run_cost is not None:
                 entry.cost = round((entry.cost or 0.0) + run_cost, 6)

backend/tests/test_console_router.py

@@ -234,6 +234,30 @@ class TestConsoleUsage:
         assert data["by_model"]["qwen"]["tokens"] == 70
         # Legacy row (empty token_usage_by_model) falls back to model_name
         assert data["by_model"]["gpt-x"]["tokens"] == 50
+        assert data["by_model"]["gpt-x"]["input_tokens"] == 50
+
+    @pytest.mark.parametrize("legacy_usage", [{}, None], ids=["empty-map", "null-map"])
+    @pytest.mark.parametrize("legacy_input_tokens", [0, 50])
+    def test_legacy_input_tokens_accumulate_with_model_usage(self, client, session_factory, legacy_usage, legacy_input_tokens):
+        async def _update_legacy_run() -> None:
+            async with session_factory() as session:
+                row = await session.get(RunRow, "r3")
+                row.model_name = "minimax-m2"
+                row.token_usage_by_model = legacy_usage
+                row.total_input_tokens = legacy_input_tokens
+                row.total_output_tokens = 50 - legacy_input_tokens
+                await session.commit()
+
+        asyncio.run(_update_legacy_run())
+        response = client.get("/api/console/usage", params={"days": 14})
+        assert response.status_code == 200
+        data = response.json()
+        model = data["by_model"]["minimax-m2"]
+        assert model["input_tokens"] == 1000 + legacy_input_tokens
+        assert model["tokens"] == 1550
+        assert model["runs"] == 3
+        assert model["cache_read_tokens"] == 500
+        assert sum(day["input_tokens"] for day in data["days"]) == sum(entry["input_tokens"] for entry in data["by_model"].values())
 
     def test_window_excludes_old_rows_but_stats_include_them(self, client):
         usage = client.get("/api/console/usage", params={"days": 7}).json()