Kimi K3 (Max) vs Mercury 2
Which is better for the money: Kimi K3 (Max) by Kimi 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
Kimi K3 (Max) scores 32.8 points higher (43.6 vs 10.8) but costs 35× as much to run the benchmark ($3658 vs $105; flagship vs mid-range cost classes). They are not alternatives at one budget: pick Kimi K3 (Max) when you need intelligence above 10.8, and Mercury 2 when 10.8 is enough.
| Metric | Kimi K3 (Max) | Mercury 2 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 43.6 | 10.8 | −75.2% Mercury 2 is 75.2% less intelligent |
| Total AA cost (lower is better) | $3,658.07 | $104.94 | −97.1% Mercury 2 is 97.1% cheaper |
| End-to-end time (lower is better) | 58.57s | 6.65s | −88.6% Mercury 2 is 88.6% faster |
| Price per 1M tokens (in / out) | $3.00 / $15.0 | $0.25 / $0.75 | |
| Input modalities | Text, Image | Text |
Mercury 2 vs Kimi K3 (Max): 75.2% less intelligence, 97.1% less cost to run the AA benchmark, and 88.6% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Kimi K3 (Max) is off the frontier: at the same or lower cost, Kimi K3.1 Pro scores 55.10 for $3117.
- Mercury 2 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (High) scores 37.80 for $94.1.
Kimi K3 (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/kimi-k3/mercury-2 · Markdown: /api/v1/compare/kimi-k3/mercury-2?format=md · llms.txt · agent skill