Single-turn model evaluation

Coding Benchmark Leaderboard

This coding benchmark is a single-turn suite with three cases: bug-fix-01, refactor-01, and migration-01. One repair is allowed. Ranking order is: scored passes desc, strict passes desc, soft passes desc, completed stages desc, repairs used asc, total cost asc, total wall time asc, then stage-1 initial passes desc. A soft pass means the case ended format-invalid, but salvaged fenced code passed.

LegendPStrict passSSoft passF:*Failed case
Showing 44 of 44
Rank
1
Model
gpt-5.3-codexopenai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
3/3
Repairs
0/3
Output
42801
Cost
$0.71
Wall
745.4
Cases
Bug FixPRefactorPMigrationP
Rank
2
Model
deepseek-v4-flashdeepseek
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
20248
Cost
$0.02
Wall
148.8
Cases
Bug FixPRefactorPMigrationP
Rank
3
Model
deepseek-v4-prodeepseek
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
20357
Cost
$0.06
Wall
290.6
Cases
Bug FixPRefactorPMigrationP
Rank
4
Model
grok-4.3grok
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8330
Cost
$0.12
Wall
90.4
Cases
Bug FixPRefactorPMigrationP
Rank
5
Model
kimi-k2.5moonshot
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
24542
Cost
$0.12
Wall
443.9
Cases
Bug FixPRefactorPMigrationP
Rank
6
Model
gpt-5.6-lunaopenai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8433
Cost
$0.13
Wall
50.3
Cases
Bug FixPRefactorPMigrationP
Rank
7
Model
qwen3.7-plusqwen
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
29815
Cost
$0.14
Wall
551.9
Cases
Bug FixPRefactorPMigrationP
Rank
8
Model
qwen3.6-plus-2026-04-02qwen
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
31973
Cost
$0.14
Wall
590.5
Cases
Bug FixPRefactorPMigrationP
Rank
9
Model
grok-build-0.1grok
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
30141
Cost
$0.14
Wall
197.8
Cases
Bug FixPRefactorPMigrationP
Rank
10
Model
glm-5-turbozai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
10931
Cost
$0.14
Wall
270.7
Cases
Bug FixPRefactorPMigrationP
Rank
11
Model
kimi-k2.7-codemoonshot
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
17069
Cost
$0.15
Wall
476.9
Cases
Bug FixPRefactorPMigrationP
Rank
12
Model
gemini-3.5-flashgoogle
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8206
Cost
$0.21
Wall
139.8
Cases
Bug FixPRefactorPMigrationP
Rank
13
Model
glm-5.2zai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
24858
Cost
$0.22
Wall
279.1
Cases
Bug FixPRefactorPMigrationP
Rank
14
Model
gpt-5.2openai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
7027
Cost
$0.24
Wall
83.0
Cases
Bug FixPRefactorPMigrationP
Rank
15
Model
grok-4.5grok
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
14539
Cost
$0.26
Wall
114.4
Cases
Bug FixPRefactorPMigrationP
Rank
16
Model
kimi-k2.6moonshot
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
50480
Cost
$0.27
Wall
811.6
Cases
Bug FixPRefactorPMigrationP
Rank
17
Model
gemini-3.1-pro-previewgoogle
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8122
Cost
$0.29
Wall
456.1
Cases
Bug FixPRefactorPMigrationP
Rank
18
Model
gemini-3-pro-previewgoogle
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8248
Cost
$0.29
Wall
460.2
Cases
Bug FixPRefactorPMigrationP
Rank
19
Model
gpt-5.2-codexopenai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
10818
Cost
$0.29
Wall
166.4
Cases
Bug FixPRefactorPMigrationP
Rank
20
Model
gpt-5.4openai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
6801
Cost
$0.31
Wall
85.9
Cases
Bug FixPRefactorPMigrationP
Rank
21
Model
kimi-k2.7-code-highspeedmoonshot
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
19211
Cost
$0.31
Wall
109.8
Cases
Bug FixPRefactorPMigrationP
Rank
22
Model
gpt-5.6-terraopenai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
7469
Cost
$0.32
Wall
49.7
Cases
Bug FixPRefactorPMigrationP
Rank
23
Model
grok-4.20-0309-reasoninggrok
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
28042
Cost
$0.33
Wall
130.9
Cases
Bug FixPRefactorPMigrationP
Rank
24
Model
qwen3.7-maxqwen
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
25116
Cost
$0.34
Wall
468.2
Cases
Bug FixPRefactorPMigrationP
Rank
25
Model
qwen3.6-max-previewqwen
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
31707
Cost
$0.40
Wall
609.3
Cases
Bug FixPRefactorPMigrationP
Rank
26
Model
kimi-k3moonshot
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
16204
Cost
$0.42
Wall
519.7
Cases
Bug FixPRefactorPMigrationP
Rank
27
Model
claude-sonnet-5anthropic
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
16263
Cost
$0.60
Wall
113.7
Cases
Bug FixPRefactorPMigrationP
Rank
28
Model
gpt-5.6-solopenai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
8322
Cost
$0.67
Wall
93.8
Cases
Bug FixPRefactorPMigrationP
Rank
29
Model
GPT-5.5openai
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
12097
Cost
$0.77
Wall
136.1
Cases
Bug FixPRefactorPMigrationP
Rank
30
Model
claude-fable-5anthropic
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
2/3
Repairs
1/3
Output
12156
Cost
$1.78
Wall
113.0
Cases
Bug FixPRefactorPMigrationP
Rank
31
Model
claude-opus-4-7anthropic
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
1/3
Repairs
2/3
Output
12714
Cost
$1.10
Wall
107.5
Cases
Bug FixPRefactorPMigrationP
Rank
32
Model
claude-opus-5anthropic
Result
3/3 Strict / 3/3 ScoredSoft 0/3 · Hard 0/3
Stages
3/3
Init
1/3
Repairs
2/3
Output
13805
Cost
$1.10
Wall
127.5
Cases
Bug FixPRefactorPMigrationP
Rank
33
Model
glm-5.1zai
Result
2/3 Strict / 3/3 ScoredSoft 1/3 · Hard 0/3
Stages
2/3
Init
2/3
Repairs
1/3
Output
13590
Cost
$0.15
Wall
241.8
Cases
Bug FixPRefactorSMigrationP
Rank
34
Model
glm-5zai
Result
2/3 Strict / 3/3 ScoredSoft 1/3 · Hard 0/3
Stages
2/3
Init
1/3
Repairs
2/3
Output
24020
Cost
$0.19
Wall
467.7
Cases
Bug FixPRefactorSMigrationP
Rank
35
Model
claude-opus-4-8anthropic
Result
2/3 Strict / 3/3 ScoredSoft 1/3 · Hard 0/3
Stages
2/3
Init
1/3
Repairs
2/3
Output
13096
Cost
$1.11
Wall
121.0
Cases
Bug FixPRefactorSMigrationP
Rank
36
Model
gemini-3.1-flash-litegoogle
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
2/3
Repairs
1/3
Output
8172
Cost
$0.04
Wall
29.9
Cases
Bug FixPRefactorF:TEST_FAILMigrationP
Rank
37
Model
grok-4.20-beta-latest-non-reasoninggrok
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
2/3
Repairs
1/3
Output
12034
Cost
$0.24
Wall
51.5
Cases
Bug FixPRefactorF:TEST_FAILMigrationP
Rank
38
Model
claude-opus-4-6anthropic
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
1/3
Repairs
1/3
Output
9034
Cost
$0.70
Wall
88.7
Cases
Bug FixF:FORMATRefactorPMigrationP
Rank
39
Model
claude-opus-4-5anthropic
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
1/3
Repairs
1/3
Output
9080
Cost
$0.70
Wall
83.6
Cases
Bug FixF:FORMATRefactorPMigrationP
Rank
40
Model
qwen3.5-plusqwen
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
1/3
Repairs
2/3
Output
7420
Cost
$0.07
Wall
82.2
Cases
Bug FixPRefactorF:TEST_FAILMigrationP
Rank
41
Model
kimi-k2-turbo-previewmoonshot
Result
2/3 Strict / 2/3 ScoredSoft 0/3 · Hard 1/3
Stages
2/3
Init
1/3
Repairs
2/3
Output
7500
Cost
$0.19
Wall
97.9
Cases
Bug FixPRefactorF:TEST_FAILMigrationP
Rank
42
Model
MiniMax-M2.7minimax
Result
0/3 Strict / 2/3 ScoredSoft 2/3 · Hard 1/3
Stages
0/3
Init
0/3
Repairs
3/3
Output
31209
Cost
$0.07
Wall
1190.0
Cases
Bug FixF:FORMATRefactorSMigrationS
Rank
43
Model
deepseek-reasonerdeepseek
Result
1/3 Strict / 1/3 ScoredSoft 0/3 · Hard 2/3
Stages
1/3
Init
1/3
Repairs
1/3
Output
13947
Cost
<$0.01
Wall
458.9
Cases
Bug FixF:FORMATRefactorF:TEST_FAILMigrationP
Rank
44
Model
MiniMax-M2.5minimax
Result
0/3 Strict / 0/3 ScoredSoft 0/3 · Hard 3/3
Stages
0/3
Init
0/3
Repairs
0/3
Output
25534
Cost
$0.05
Wall
1135.2
Cases
Bug FixF:FORMATRefactorF:FORMATMigrationF:FORMAT

Numeric columns are sortable. Currency is rounded to two decimals for readability.