DeepSeek V4.1 Flash (Max) vs Mercury 2
Which is better for the money: DeepSeek V4.1 Flash (Max) by DeepSeek or Mercury 2 by Inception? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Trade-off
Verdict
Depends on your budget
DeepSeek V4.1 Flash (Max) scores 28.7 points higher (39.5 vs 10.8) but costs 6.4× as much to run the benchmark ($672 vs $105; premium vs mid-range cost classes). They are not alternatives at one budget: pick DeepSeek V4.1 Flash (Max) when you need intelligence above 10.8, and Mercury 2 when 10.8 is enough.
| Metric | DeepSeek V4.1 Flash (Max) | Mercury 2 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 39.5 | 10.8 | −72.7% Mercury 2 is 72.7% less intelligent |
| Total AA cost (lower is better) | $671.96 | $104.94 | −84.4% Mercury 2 is 84.4% cheaper |
| End-to-end time (lower is better) | 15.00s | 6.65s | −55.7% Mercury 2 is 55.7% faster |
| Price per 1M tokens (in / out) | $0.29 / $1.20 | $0.25 / $0.75 | |
| Input modalities | Text, Image | Text |
Mercury 2 vs DeepSeek V4.1 Flash (Max): 72.7% less intelligence, 84.4% less cost to run the AA benchmark, and 55.7% faster end-to-end.
Today's Pareto frontier
Against the frontier
- DeepSeek V4.1 Flash (Max) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Xhigh) scores 51.00 for $662.
- Mercury 2 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
DeepSeek V4.1 Flash (Max)Mercury 2Today's Pareto frontierOpen-weights frontierEvery other model
Matchup
Compare something else
Agents:
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/deepseek-v4-1-flash/mercury-2 · Markdown: /api/v1/compare/deepseek-v4-1-flash/mercury-2?format=md · llms.txt · agent skill