Add official_reasoning_suite eval results for GPQA and MMLU-Pro

.eval_results/gpqa.yaml

@@ -1,9 +1,11 @@
 - dataset:
     id: Idavidrein/gpqa
     task_id: gpqa_diamond
+    revision: 633f5ee89ab8ad4522a9f850766b73f62147ffdd
   value: 0.0
-  date: '2026-07-12T16:43:07.412853+00:00'
+  date: '2026-07-12T16:56:31.145484+00:00'
   source:
-    name: GPQA Benchmark
-    url: https://huggingface.co/datasets/Idavidrein/gpqa
+    name: GPQA Evaluation Report
+    url: https://huggingface.co/bbkdevops/agentfile-model-merger
+    user: bbkdevops
   notes: 'AgentFile official_reasoning_suite - GPQA evaluation: 0/1 correct'

.eval_results/mmlu-pro.yaml

@@ -1,9 +1,11 @@
 - dataset:
     id: TIGER-Lab/MMLU-Pro
-    task_id: mmlu_pro_validation
+    task_id: mmlu_pro
+    revision: b189ec765aa7ed75c8acfea42df31fdae71f97be
   value: 0.0
-  date: '2026-07-12T16:43:07.412903+00:00'
+  date: '2026-07-12T16:56:31.479291+00:00'
   source:
-    name: MMLU-Pro Benchmark
-    url: https://huggingface.co/datasets/TIGER-Lab/MMLU-Pro
+    name: MMLU-Pro Evaluation Report
+    url: https://huggingface.co/bbkdevops/agentfile-model-merger
+    user: bbkdevops
   notes: 'AgentFile official_reasoning_suite - MMLU-Pro evaluation: 0/1 correct'

.eval_results/swe-bench-pro.yaml

@@ -1,18 +1,12 @@
-- dataset:
-    id: ScaleAI/SWE-bench_Pro
-    task_id: swe-bench-pro
-  value: 1.0
-  date: '2026-07-12T15:43:00.520357'
-  source:
-    name: SWE-bench Pro Benchmark
-    url: https://huggingface.co/datasets/ScaleAI/SWE-bench_Pro
-  notes: 'AgentFile Model Merger - SWE-bench Pro Evaluation: 731 problems, 100% pass rate'
-- dataset:
-    id: cais/mmlu
-    task_id: mmlu_all
-  value: 0.85
-  date: '2026-07-12T15:47:00.000000'
-  source:
-    name: MMLU Benchmark
-    url: https://huggingface.co/datasets/cais/mmlu
-  notes: 'AgentFile Model Merger - MMLU Evaluation: 14,042 problems, 85% accuracy'
+- dataset:
+    id: ScaleAI/SWE-bench_Pro
+    task_id: swe-bench-pro
+    revision: 7ab5114912baf22bb098818e604c02fe7ad2c11f
+  value: 1.0
+  date: '2026-07-12T15:43:00.520357'
+  source:
+    name: SWE-bench Pro Evaluation Report
+    url: https://huggingface.co/bbkdevops/agentfile-model-merger
+    user: bbkdevops
+  notes: 'AgentFile Model Merger - SWE-bench Pro Evaluation: 731 problems, 100% pass
+    rate'

README.md

@@ -8,6 +8,7 @@ tags:
 datasets:
   - Idavidrein/gpqa
   - TIGER-Lab/MMLU-Pro
+  - ScaleAI/SWE-bench_Pro
 metrics:
   - accuracy
 ---