nm-research commited on
Commit
f406783
·
verified ·
1 Parent(s): 47fa8d9

Update SWE-Bench Lite eval results (official harness resolved-rate + submission-rate, vLLM 0.24.0)

Browse files
every_eval_ever/swebench_lite_dev.json CHANGED
@@ -1,19 +1,19 @@
1
  {
2
  "schema_version": "0.2.2",
3
- "evaluation_id": "swebench_lite_dev/RedHatAI/Kimi-K2.6-NVFP4/1782913700",
4
- "retrieved_timestamp": "1782913700",
5
  "source_metadata": {
6
- "source_name": "mini-swe-agent",
7
  "source_type": "evaluation_run",
8
  "source_organization_name": "RedHatAI",
9
- "evaluator_relationship": "third_party",
10
  "additional_details": {
11
- "note": "submission-rate aggregation across reruns"
12
  }
13
  },
14
  "eval_library": {
15
- "name": "mini-swe-agent",
16
- "version": "2.4.3"
17
  },
18
  "model_info": {
19
  "name": "RedHatAI/Kimi-K2.6-NVFP4",
@@ -21,7 +21,7 @@
21
  "developer": "RedHatAI",
22
  "inference_engine": {
23
  "name": "vllm",
24
- "version": "0.22.1"
25
  },
26
  "additional_details": {
27
  "profile": "kimi_nvfp4",
@@ -31,14 +31,33 @@
31
  },
32
  "evaluation_results": [
33
  {
34
- "evaluation_name": "swebench_lite_dev/submission_rate",
35
  "source_data": {
36
  "dataset_name": "SWE-bench Lite dev",
37
  "source_type": "hf_dataset",
38
  "hf_repo": "princeton-nlp/SWE-Bench_Lite",
39
  "hf_split": "dev",
40
- "samples_number": 23,
41
- "sample_ids": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
42
  "marshmallow-code__marshmallow-1343",
43
  "marshmallow-code__marshmallow-1359",
44
  "pvlib__pvlib-python-1072",
@@ -53,35 +72,13 @@
53
  "pydicom__pydicom-901",
54
  "pylint-dev__astroid-1196",
55
  "pylint-dev__astroid-1268",
56
- "pylint-dev__astroid-1333",
57
  "pylint-dev__astroid-1866",
58
  "pylint-dev__astroid-1978",
59
  "pyvista__pyvista-4315",
60
  "sqlfluff__sqlfluff-1517",
61
- "sqlfluff__sqlfluff-1625",
62
- "sqlfluff__sqlfluff-1733",
63
- "sqlfluff__sqlfluff-1763",
64
  "sqlfluff__sqlfluff-2419"
65
- ]
66
- },
67
- "metric_config": {
68
- "evaluation_description": "Submission rate (Submitted / total instances).",
69
- "lower_is_better": false,
70
- "score_type": "continuous",
71
- "min_score": 0.0,
72
- "max_score": 1.0
73
- },
74
- "score_details": {
75
- "score": 0.9130434782608695,
76
- "details": {
77
- "total_instances": "23",
78
- "submitted_instances": "21",
79
- "unsubmitted_instances": "2",
80
- "status_counts": "{\"RepeatedFormatError\": 2, \"Submitted\": 21}",
81
- "unsubmitted_ids": "[\"sqlfluff__sqlfluff-1625\", \"sqlfluff__sqlfluff-1733\"]",
82
- "source_exit_status_files": "[\"/home/shubhra/kimik2.6_evals/runs/swebench/20260629T171225Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782760165.6381476.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T135435Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782827902.5985954.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T141107Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782830121.1086848.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T151900Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782833877.9161725.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T190648Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782846629.8118455.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T192452Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782848479.344266.yaml\", \"/home/shubhra/kimik2.6_evals/runs/swebench/20260630T195651Z_kimi_nvfp4_lite_dev/output/exit_statuses_1782850223.0775344.yaml\"]"
83
- },
84
- "completed_ids": [
85
  "marshmallow-code__marshmallow-1343",
86
  "marshmallow-code__marshmallow-1359",
87
  "pvlib__pvlib-python-1072",
@@ -106,6 +103,50 @@
106
  "sqlfluff__sqlfluff-1763",
107
  "sqlfluff__sqlfluff-2419"
108
  ],
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
109
  "submitted_ids": [
110
  "marshmallow-code__marshmallow-1343",
111
  "marshmallow-code__marshmallow-1359",
@@ -126,6 +167,8 @@
126
  "pylint-dev__astroid-1978",
127
  "pyvista__pyvista-4315",
128
  "sqlfluff__sqlfluff-1517",
 
 
129
  "sqlfluff__sqlfluff-1763",
130
  "sqlfluff__sqlfluff-2419"
131
  ]
@@ -144,4 +187,4 @@
144
  }
145
  }
146
  ]
147
- }
 
1
  {
2
  "schema_version": "0.2.2",
3
+ "evaluation_id": "swebench_lite_dev/RedHatAI/Kimi-K2.6-NVFP4/1784247001.8595154",
4
+ "retrieved_timestamp": "1784247001.8595154",
5
  "source_metadata": {
6
+ "source_name": "SWE-bench harness",
7
  "source_type": "evaluation_run",
8
  "source_organization_name": "RedHatAI",
9
+ "evaluator_relationship": "first_party",
10
  "additional_details": {
11
+ "note": "predictions generated with mini-swe-agent, resolved-rate scored with the official SWE-bench harness (swebench.harness.run_evaluation)"
12
  }
13
  },
14
  "eval_library": {
15
+ "name": "swebench",
16
+ "version": "4.1.0"
17
  },
18
  "model_info": {
19
  "name": "RedHatAI/Kimi-K2.6-NVFP4",
 
21
  "developer": "RedHatAI",
22
  "inference_engine": {
23
  "name": "vllm",
24
+ "version": "0.24.0"
25
  },
26
  "additional_details": {
27
  "profile": "kimi_nvfp4",
 
31
  },
32
  "evaluation_results": [
33
  {
34
+ "evaluation_name": "swebench_lite_dev/resolved_rate",
35
  "source_data": {
36
  "dataset_name": "SWE-bench Lite dev",
37
  "source_type": "hf_dataset",
38
  "hf_repo": "princeton-nlp/SWE-Bench_Lite",
39
  "hf_split": "dev",
40
+ "samples_number": 23
41
+ },
42
+ "metric_config": {
43
+ "evaluation_description": "Resolve rate: fraction of benchmark instances where the submitted patch passes verification (0.0-1.0), scored with the official SWE-bench harness.",
44
+ "lower_is_better": false,
45
+ "score_type": "continuous",
46
+ "min_score": 0.0,
47
+ "max_score": 1.0
48
+ },
49
+ "score_details": {
50
+ "score": 0.21739130434782608,
51
+ "details": {
52
+ "total_instances": "23",
53
+ "resolved_instances": "5",
54
+ "submitted_instances": "23",
55
+ "completed_instances": "19",
56
+ "unresolved_instances": "14",
57
+ "empty_patch_instances": "4",
58
+ "error_instances": "0"
59
+ },
60
+ "completed_ids": [
61
  "marshmallow-code__marshmallow-1343",
62
  "marshmallow-code__marshmallow-1359",
63
  "pvlib__pvlib-python-1072",
 
72
  "pydicom__pydicom-901",
73
  "pylint-dev__astroid-1196",
74
  "pylint-dev__astroid-1268",
 
75
  "pylint-dev__astroid-1866",
76
  "pylint-dev__astroid-1978",
77
  "pyvista__pyvista-4315",
78
  "sqlfluff__sqlfluff-1517",
 
 
 
79
  "sqlfluff__sqlfluff-2419"
80
+ ],
81
+ "submitted_ids": [
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
82
  "marshmallow-code__marshmallow-1343",
83
  "marshmallow-code__marshmallow-1359",
84
  "pvlib__pvlib-python-1072",
 
103
  "sqlfluff__sqlfluff-1763",
104
  "sqlfluff__sqlfluff-2419"
105
  ],
106
+ "resolved_ids": [
107
+ "marshmallow-code__marshmallow-1343",
108
+ "pydicom__pydicom-1256",
109
+ "pydicom__pydicom-1694",
110
+ "pylint-dev__astroid-1268",
111
+ "sqlfluff__sqlfluff-2419"
112
+ ]
113
+ },
114
+ "generation_config": {
115
+ "generation_args": {
116
+ "agentic_eval_config": {
117
+ "available_tools": [
118
+ {
119
+ "name": "bash"
120
+ }
121
+ ]
122
+ },
123
+ "max_attempts": 1
124
+ }
125
+ }
126
+ },
127
+ {
128
+ "evaluation_name": "swebench_lite_dev/submission_rate",
129
+ "source_data": {
130
+ "dataset_name": "SWE-bench Lite dev",
131
+ "source_type": "hf_dataset",
132
+ "hf_repo": "princeton-nlp/SWE-Bench_Lite",
133
+ "hf_split": "dev",
134
+ "samples_number": 23
135
+ },
136
+ "metric_config": {
137
+ "evaluation_description": "Submission rate (Submitted / total instances).",
138
+ "lower_is_better": false,
139
+ "score_type": "continuous",
140
+ "min_score": 0.0,
141
+ "max_score": 1.0
142
+ },
143
+ "score_details": {
144
+ "score": 1.0,
145
+ "details": {
146
+ "total_instances": "23",
147
+ "submitted_instances": "23",
148
+ "unsubmitted_instances": "0"
149
+ },
150
  "submitted_ids": [
151
  "marshmallow-code__marshmallow-1343",
152
  "marshmallow-code__marshmallow-1359",
 
167
  "pylint-dev__astroid-1978",
168
  "pyvista__pyvista-4315",
169
  "sqlfluff__sqlfluff-1517",
170
+ "sqlfluff__sqlfluff-1625",
171
+ "sqlfluff__sqlfluff-1733",
172
  "sqlfluff__sqlfluff-1763",
173
  "sqlfluff__sqlfluff-2419"
174
  ]
 
187
  }
188
  }
189
  ]
190
+ }