modelfit_chat returns ok

main.py

@@ -188,7 +188,7 @@ async def _fetch_url_to_dataframe(data_url: str) -> pd.DataFrame:
     return df
 
 
-async def _forward(path: str, method: str = "GET", json_body=None, user_token: str | None = None):
+async def _forward(path: str, method: str = "GET", json_body=None, user_token: str | None = None, success_response_override: dict | None = None):
     """
     Forward request to the PRIVATE Space:
       - 'Authorization: Bearer <HF_TOKEN>' to pass HF private gate
@@ -269,6 +269,10 @@ async def _forward(path: str, method: str = "GET", json_body=None, user_token: s
     if r is None:
         raise RuntimeError("Upstream request did not complete")
 
+    # When caller wants a minimal response on success (e.g. for chat to avoid response size limits)
+    if success_response_override is not None and r.status_code == 200:
+        return JSONResponse(status_code=200, content=success_response_override)
+
     ct = r.headers.get("content-type", "")
 
     if "application/json" in ct:
@@ -586,7 +590,8 @@ async def modelfit_chat(
     """
     Fetch training data from data_url (HTTPS), infer schema from column order, then call /modelfit/ on the API.
     Column order: 1st = id_col, 2nd = time_col, 3rd..second-to-last = current_features, last = y.
-    All other parameters use defaults. Returns JSON fit result.
+    All other parameters use defaults. On success returns a minimal JSON (ok, message) to stay under
+    chat platform response size limits; errors are returned in full.
     """
     user_token = _extract_user_token(req)
     if not user_token:
@@ -614,7 +619,13 @@ async def modelfit_chat(
         "meanvar_test": False,
         "signif": 0.05,
     }
-    return await _forward("/modelfit/", "POST", json_body=payload, user_token=user_token)
+    return await _forward(
+        "/modelfit/",
+        "POST",
+        json_body=payload,
+        user_token=user_token,
+        success_response_override={"ok": True, "message": "Model fitted. Call modelforecast_chat with your forecast data URL next."},
+    )
 
 
 @app.post("/modelforecast_chat/")