Claude Fable 5 vs DeepSeek V4: GA Frontier vs Open-Weights Frontier

Claude Fable 5DeepSeek V4
arenaElo14921395
humanEval9891
sweBench7762
mmlu95.191.5
gpqa9273
aime9490
Price in/out ($/M)$10 / $50$1.74 / $3.48

Our verdict

DeepSeek V4 is the open-weights flagship: MIT-licensed, 1M context, self-hostable and roughly 6× cheaper per token. Fable 5 keeps the capability lead on long-horizon agents and frontier reasoning, but at a premium and behind 30-day data retention. Pick V4 for teams that must run frontier-tier intelligence on their own hardware or under data-residency rules; pick Fable 5 when maximum capability and Anthropic safety tuning justify the cost.

Claude Fable 5 · DeepSeek V4 · All AI model comparisons