Claude Opus 4.8 vs DeepSeek V4 Pro
8 shared benchmarks
8
wins
0
ties
0
wins
39.4%
APEX-Agents
38.3%
84.3%
BrowseComp
83.4%
91.0%
GPQA Diamond
90.1%
96.7%
HMMT Feb 2026
95.2%
49.8%
Humanity's Last Exam
37.7%
77.8%
MCP-Atlas
73.6%
69.2%
SWE-bench Pro
55.4%
59.9%
Toolathlon
51.8%