diff --git a/docs/devir-notlari.md b/docs/devir-notlari.md
index ac90f93..54ca1fe 100644
--- a/docs/devir-notlari.md
+++ b/docs/devir-notlari.md
@@ -45,7 +45,7 @@ After the jobs:
 | G0170 (H): DeepSeek did not stop; it assumed the rounding (gap: rounding target and mode of the loyalty discount). Gap maybe not critical. | Waiting for the DeepSeek rerun with budget 60 (in W7A). If it again implements: suggest a stronger gap, as for G0174. G0170 is in the stage 1 subset. |
 | Second model for the empirical filter (easy tasks) | Qwen baseline gives 25 tasks only. Decide later if more models run on all 109 tasks. |
 | EPOD server bug: a second write of a PROG/FUNC is not activated (`outcome: notExecuted`), also not by `sap_activate`. | Workaround in the proxy (ADT REST lock/write/activate). The server fix is the other project's work. |
-| Separate SAP user for the harness (dumps under user KESELI) | Proposed, no decision. |
+| Separate SAP user for the harness (dumps under user KESELI) | Decided 2026-10-03 (Kral): not needed. |
 | G0122 (C FUNC): not accepted after 4 generations | Left out. C has 14 tasks (target 15). |
 
 ## 3. Pending follow-ups
diff --git a/tasks_gen/eval/G0173/empirical.json b/tasks_gen/eval/G0173/empirical.json
index 1d532d7..eafb2fa 100644
--- a/tasks_gen/eval/G0173/empirical.json
+++ b/tasks_gen/eval/G0173/empirical.json
@@ -28,5 +28,33 @@
    "seconds": 66.6,
    "run_dir": "runs/emp/12015_G0173_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "keywords",
+    "group": [
+     [
+      "18",
+      "15"
+     ],
+     [
+      "minimum",
+      "20"
+     ],
+     [
+      "contradict",
+      "rule"
+     ]
+    ]
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 18,
+  "seconds": 164.6,
+  "run_dir": "runs/emp/18007_G0173_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file
diff --git a/tasks_gen/eval/G0175/empirical.json b/tasks_gen/eval/G0175/empirical.json
index 206bca6..4fb96f5 100644
--- a/tasks_gen/eval/G0175/empirical.json
+++ b/tasks_gen/eval/G0175/empirical.json
@@ -17,5 +17,22 @@
    "seconds": 7.9,
    "run_dir": "runs/emp/12013_G0175_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "judge",
+    "model": "deepseek-v4.1-flash:cloud",
+    "names_gap": true,
+    "reason": "The report explicitly identifies that rule 1 requires HARVEST_DATE >= P_FROM while rule 5 requires HARVEST_DATE < P_FROM, making the predicates mutually exclusive and the result set empty."
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 3,
+  "seconds": 9.2,
+  "run_dir": "runs/emp/18008_G0175_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file
diff --git a/tasks_gen/eval/G0177/empirical.json b/tasks_gen/eval/G0177/empirical.json
index 24d71e2..a9511a5 100644
--- a/tasks_gen/eval/G0177/empirical.json
+++ b/tasks_gen/eval/G0177/empirical.json
@@ -28,5 +28,33 @@
    "seconds": 12.0,
    "run_dir": "runs/emp/12011_G0177_llm_deepseek-v4.1-flash_cloud"
   }
+ },
+ "deepseek-v4.1-flash:cloud": {
+  "score": 100,
+  "parts": {
+   "total": 100,
+   "stop": "stopped",
+   "gap": {
+    "how": "keywords",
+    "group": [
+     [
+      "contradiction",
+      "rules"
+     ],
+     [
+      "contradict",
+      "rules"
+     ],
+     [
+      "confirmed",
+      "cancelled"
+     ]
+    ]
+   }
+  },
+  "hidden": "0/0",
+  "tool_calls": 8,
+  "seconds": 13.7,
+  "run_dir": "runs/emp/18009_G0177_llm_deepseek-v4.1-flash_cloud"
  }
 }
\ No newline at end of file