Claude Fable 5 vs Claude Opus 4.8
26 shared benchmarks
24
wins
1
ties
1
wins
43.3%
APEX-Agents
39.4%
29.1%
AutomationBench
27.2%
90.5%
BabyVision
81.2%
88.0%
BrowseComp
84.3%
88.9%
CharXiv
80.5%
70.0%
DeepSWE
59.0%
86.6%
FrontierSWE
66.7%
92.6%
GPQA Diamond
91.0%
53.3%
Humanity's Last Exam
49.8%
57.4%
JobBench
48.4%
94.8%
MATH-Vision
86.7%
84.7%
MCP-Atlas
77.8%
49.9%
MLS Bench
42.8%
81.2%
MMMU Pro
78.9%
69.9%
OfficeQA Pro
63.9%
89.8%
OmniDocBench
87.9%
57.2%
PerceptionBench
47.2%
41.4%
PostTrain Bench
34.1%
76.8%
Program Bench
71.9%
35.0%
SWE Marathon
40.0%
80.3%
SWE-bench Pro
69.2%
34.7%
SpreadsheetBench 2
31.55%
84.6%
Terminal-Bench 2.1
84.6%
77.9%
Toolathlon-Verified
76.2%
56.7%
WorldVQA
39.1%
23.0%
ZeroBench
17.0%