DeepSeek V4 vs Claude Opus 4.7: Open-Source Frontier Meets the Top Closed Model

DeepSeek V4Claude Opus 4.7
arenaElo13951420
humanEval9193.1
sweBench6264.3
mmlu91.592.3
gpqa7365.2
aime9088
Price in/out ($/M)$1.74 / $3.48$5 / $25

Our verdict

DeepSeek V4 (released April 2026) matches the 1M-token context of Opus 4.7 at roughly 1/6 the price ($1.74/$3.48 vs $15/$75 per M tokens). Opus 4.7 still leads on Chatbot Arena, agentic tool use and coding polish; V4 wins on cost-per-intelligence and is fully self-hostable under MIT license. Pick Opus 4.7 for mission-critical agents, V4 for scale-sensitive STEM and document workloads.

DeepSeek V4 · Claude Opus 4.7 · All AI model comparisons