GPT-5.6 Sol
OpenAI
Parameters
—
Architecture
—
Released
—
License
—
About
No detailed description available for this model.
Benchmark Scores
| Benchmark | Score | Date | Standing |
|---|---|---|---|
|
Terminal-Bench 4.0
coding_agent
|
51.41%
|
19.10.2026 |
#5/10
better than 56%
|
|
Terminal Bench 2.1
coding_agent
|
98.00%
|
18.08.2026 |
#2/44
better than 98%
|
|
MLS-Bench-Lite
coding_agent
|
84.05%
|
18.08.2026 |
#3/9
better than 75%
|
|
PaperBench
coding_agent
|
95.99%
|
18.08.2026 |
#2/10
better than 89%
|
|
AndroidBench
coding_agent
|
62.50%
|
18.08.2026 |
#3/5
better than 50%
|
|
QwenSWEBench
coding_agent
|
65.41%
|
18.08.2026 |
#5/9
better than 50%
|
|
QwenQoderBench
coding_agent
|
64.64%
|
18.08.2026 |
#4/5
better than 25%
|
|
QwenReactBench
coding_agent
|
11.21%
|
18.08.2026 |
#4/5
better than 25%
|
|
QwenSVGBench
coding_agent
|
100.00%
|
18.08.2026 |
#1/5
better than 100%
|
|
CoWorkBench
general_agent
|
85.71%
|
18.08.2026 |
#5/12
better than 64%
|
|
JobBench
general_agent
|
51.08%
|
18.08.2026 |
#11/19
better than 44%
|
|
SkillsBench
general_agent
|
100.00%
|
18.08.2026 |
#1/9
better than 100%
|
|
Toolathlon Verified
general_agent
|
94.01%
|
18.08.2026 |
#7/35
better than 82%
|
|
GPQA Diamond
stem_reasoning
|
99.62%
|
18.08.2026 |
#3/80
better than 97%
|
|
Humanity's Last Exam
stem_reasoning
|
85.42%
|
18.08.2026 |
#5/71
better than 94%
|
|
IFBench
instruction_following
|
81.30%
|
18.08.2026 |
#24/42
better than 44%
|
|
$OneMillion-Bench
general_capabilities
|
45.63%
|
18.08.2026 |
#7/11
better than 40%
|
|
HealthBench
general_capabilities
|
72.32%
|
18.08.2026 |
#3/6
better than 60%
|
|
PLawBench
general_capabilities
|
93.71%
|
18.08.2026 |
#2/5
better than 75%
|
|
PRBench-Legal
general_capabilities
|
100.00%
|
18.08.2026 |
#1/5
better than 50%
|
|
PRBench-Finance
general_capabilities
|
75.65%
|
18.08.2026 |
#3/5
better than 50%
|
|
MRCR v2 256K (8-needle)
long_context
|
100.00%
|
18.08.2026 |
#1/4
better than 100%
|
|
LongBench v2
long_context
|
95.48%
|
18.08.2026 |
#2/22
better than 95%
|
|
Agents' Last Exam
general_agent
|
61.35%
|
18.08.2026 |
#4/29
better than 89%
|
|
OfficeQA Pro
general_agent
|
95.94%
|
28.08.2026 |
#5/14
better than 54%
|
|
Apex-Agents
general_agent
|
53.43%
|
28.08.2026 |
#6/24
better than 78%
|
|
HLE (with tools)
stem_reasoning
|
78.61%
|
18.08.2026 |
#10/49
better than 81%
|
|
SWE-Marathon
coding_agent
|
81.85%
|
28.08.2026 |
#5/17
better than 69%
|
|
GDP.pdf
long_context
|
76.13%
|
19.10.2026 |
#2/13
better than 92%
|
|
Terminal-Bench 3.0
coding_agent
|
100.00%
|
28.08.2026 |
#1/7
better than 100%
|
|
PostTrainBench
coding_agent
|
78.84%
|
28.08.2026 |
#10/22
better than 57%
|
|
Cybergym
general_agent
|
90.89%
|
28.08.2026 |
#3/17
better than 88%
|
|
ExploitGym (2h)
cybersecurity
|
100.00%
|
28.08.2026 |
#1/7
better than 100%
|
|
ExploitGym (6h)
cybersecurity
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
ExploitBench
cybersecurity
|
97.20%
|
28.08.2026 |
#2/8
better than 86%
|
|
GDPVal-AA v2
general_agent
|
94.48%
|
28.08.2026 |
#6/38
better than 86%
|
|
SWE-bench Multilingual
coding_agent
|
81.78%
|
28.08.2026 |
#19/48
better than 62%
|
|
DeepSWE
coding_agent
|
100.00%
|
28.08.2026 |
#1/21
better than 100%
|
|
SWE Atlas - QnA
coding_agent
|
89.23%
|
28.08.2026 |
#2/24
better than 96%
|
|
SWE Atlas - TW
coding_agent
|
70.30%
|
28.08.2026 |
#4/20
better than 79%
|
|
SWE Atlas - RF
coding_agent
|
86.36%
|
28.08.2026 |
#3/14
better than 85%
|
|
Terminal-Bench 2.1 (Terminus-2)
coding_agent
|
100.00%
|
28.08.2026 |
#1/44
better than 100%
|
|
NL2Repo
coding_agent
|
71.54%
|
28.08.2026 |
#10/41
better than 78%
|
|
Harbor-Index
coding_agent
|
74.33%
|
28.08.2026 |
#2/7
better than 83%
|
|
Hy-Backend 2.0 (Internal)
coding_agent
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
Hy-SWE Max Verified (Internal)
coding_agent
|
98.58%
|
28.08.2026 |
#2/8
better than 86%
|
|
Hy-CompanyBench V2 (Internal)
general_agent
|
95.10%
|
28.08.2026 |
#2/8
better than 86%
|
|
WideSearch
general_agent
|
100.00%
|
28.08.2026 |
#1/24
better than 100%
|
|
DRACO
general_agent
|
53.42%
|
28.08.2026 |
#3/12
better than 73%
|
|
Hy-LifeSearch (Internal)
general_agent
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
Hy-BrowseComp-Pro2 (Internal)
general_agent
|
66.89%
|
28.08.2026 |
#3/8
better than 71%
|
|
SkillsBench Avg5
coding_agent
|
93.26%
|
28.08.2026 |
#6/16
better than 67%
|
|
BankerToolBench
general_agent
|
83.89%
|
28.08.2026 |
#3/15
better than 86%
|
|
E-Bench (Internal)
general_agent
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
E-Bench-Code (Internal)
coding_agent
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
Hy-FinAgentBench (Internal)
domain_finance
|
100.00%
|
28.08.2026 |
#1/8
better than 100%
|
|
Hy-FinmodelBench v2 (Internal)
domain_finance
|
98.13%
|
28.08.2026 |
#2/8
better than 86%
|
|
BioMysteryBench
stem_reasoning
|
70.82%
|
28.08.2026 |
#3/10
better than 78%
|
|
CritPt (no tools)
stem_reasoning
|
100.00%
|
28.08.2026 |
#1/21
better than 100%
|
|
SUPERChem
stem_reasoning
|
87.14%
|
28.08.2026 |
#2/8
better than 86%
|
|
HorizonMath (pass@4)
stem_reasoning
|
100.00%
|
28.08.2026 |
#1/7
better than 100%
|
|
MathArena Apex 2025
stem_reasoning
|
97.34%
|
28.08.2026 |
#2/8
better than 86%
|
|
BrokenArXiv
stem_reasoning
|
73.92%
|
28.08.2026 |
#2/7
better than 83%
|
|
WorkSpaceBench
general_agent
|
44.05%
|
18.08.2026 |
#6/11
better than 50%
|
|
ProgramBench
coding_agent
|
29.03%
|
28.08.2026 |
#10/18
better than 47%
|
|
ArXivMath
stem_reasoning
|
100.00%
|
28.08.2026 |
#1/7
better than 100%
|
|
MCP-Atlas
general_agent
|
91.25%
|
28.08.2026 |
#5/54
better than 92%
|
|
DeepSWE 1.1
coding_agent
|
94.79%
|
18.08.2026 |
#6/34
better than 85%
|
|
CharXiv (RQ)
document_understanding
|
75.89%
|
19.10.2026 |
#4/34
better than 91%
|
|
LOCA-Bench (256k)
long_context
|
97.27%
|
19.10.2026 |
#2/7
better than 83%
|
|
SWE Refactor Bench
coding_agent
|
87.86%
|
19.10.2026 |
#2/6
better than 80%
|
|
NL2Repo-Bench
coding_agent
|
65.55%
|
19.10.2026 |
#8/23
better than 68%
|
|
Automation-Bench
general_agent
|
36.35%
|
18.08.2026 |
#15/27
better than 46%
|
|
SWE-bench Pro
coding_agent
|
72.02%
|
18.08.2026 |
#8/77
better than 89%
|
|
OSWorld 2.0 (Binary)
general_agent
|
92.77%
|
19.10.2026 |
#2/8
better than 86%
|
|
BabyVision
vision_language
|
93.62%
|
19.10.2026 |
#5/24
better than 83%
|
What is a good Terminal-Bench 4.0 score? What is a good Terminal Bench 2.1 score? What is a good MLS-Bench-Lite score?
No comments yet
Be the first to share your thoughts!