Claude Opus 4.7
Anthropic
Parameters
—
Architecture
—
Released
—
License
—
About
No detailed description available for this model.
Benchmark Scores
| Benchmark | Score | Date | Standing |
|---|---|---|---|
|
CL-bench
coding_agent
|
75.00%
|
11.06.2026 |
#2/5
better than 75%
|
|
VIBE-V2
coding_agent
|
100.00%
|
11.06.2026 |
#1/9
better than 100%
|
|
SWE-bench Verified
coding_agent
|
100.00%
|
11.06.2026 |
#1/60
better than 100%
|
|
Terminal Bench 2.1
coding_agent
|
72.84%
|
11.06.2026 |
#24/44
better than 47%
|
|
SWE Atlas - QnA
coding_agent
|
65.69%
|
11.06.2026 |
#10/24
better than 61%
|
|
SWE Atlas - TW
coding_agent
|
52.57%
|
11.06.2026 |
#8/20
better than 63%
|
|
SWE-efficiency
coding_agent
|
86.50%
|
11.06.2026 |
#2/5
better than 75%
|
|
LiveSQLBench
coding_agent
|
100.00%
|
11.06.2026 |
#1/5
better than 100%
|
|
SVG-Bench
coding_agent
|
88.82%
|
11.06.2026 |
#3/9
better than 75%
|
|
PostTrainBench
coding_agent
|
100.00%
|
11.06.2026 |
#1/22
better than 100%
|
|
PaperBench
coding_agent
|
44.71%
|
11.06.2026 |
#6/10
better than 44%
|
|
DRACO
general_agent
|
53.42%
|
11.06.2026 |
#3/12
better than 73%
|
|
BankerToolBench
general_agent
|
96.67%
|
11.06.2026 |
#2/15
better than 93%
|
|
SpreadSheetBench-v1
general_agent
|
98.35%
|
11.06.2026 |
#2/9
better than 88%
|
|
YC-Bench
general_agent
|
100.00%
|
11.06.2026 |
#1/8
better than 100%
|
|
LOCA-Bench (256k)
long_context
|
70.00%
|
11.06.2026 |
#3/7
better than 67%
|
|
MCP-Atlas
general_agent
|
84.07%
|
11.06.2026 |
#10/54
better than 83%
|
|
Claw-Eval Avg
coding_agent
|
86.45%
|
11.06.2026 |
#8/40
better than 82%
|
|
OSWorld-Verified
general_agent
|
95.94%
|
11.06.2026 |
#3/25
better than 92%
|
|
OmniDocBench 1.5
document_understanding
|
97.48%
|
11.06.2026 |
#7/26
better than 72%
|
|
IMO 2025
stem_reasoning
|
17.40
|
11.06.2026 |
#4/4
better than 0%
|
|
OfficeQA Pro
general_agent
|
51.72%
|
11.06.2026 |
#13/14
better than 8%
|
|
MMMU-Pro
vision_language
|
87.70%
|
11.06.2026 |
#11/35
better than 71%
|
|
Apex-Agents
general_agent
|
49.22%
|
11.06.2026 |
#9/24
better than 65%
|
|
BrowseComp
general_agent
|
85.62%
|
11.06.2026 |
#12/45
better than 73%
|
|
USAMO 2026
stem_reasoning
|
52.80
|
11.06.2026 |
#4/4
better than 0%
|
|
KernelBench Hard
coding_agent
|
100.00%
|
11.06.2026 |
#1/5
better than 100%
|
|
GDPVal
general_agent
|
90.04%
|
11.06.2026 |
#3/20
better than 89%
|
|
NL2Repo-Bench
coding_agent
|
64.62%
|
11.06.2026 |
#9/23
better than 64%
|
|
SWE-bench Pro
coding_agent
|
71.68%
|
11.06.2026 |
#10/77
better than 88%
|
|
VideoMMMU
video_understanding
|
87.21%
|
11.06.2026 |
#8/19
better than 61%
|
What is a good CL-bench score? What is a good VIBE-V2 score? What is a good SWE-bench Verified score?
No comments yet
Be the first to share your thoughts!