Kimi K3 (Low) vs Mercury 2
Which is better for the money: Kimi K3 (Low) 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 (Low) scores 19.3 points higher (30.1 vs 10.8) but costs 11× as much to run the benchmark ($1190 vs $105; premium vs mid-range cost classes). They are not alternatives at one budget: pick Kimi K3 (Low) when you need intelligence above 10.8, and Mercury 2 when 10.8 is enough.
| Metric | Kimi K3 (Low) | Mercury 2 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 30.1 | 10.8 | −64.1% Mercury 2 is 64.1% less intelligent |
| Total AA cost (lower is better) | $1,190.42 | $104.94 | −91.2% Mercury 2 is 91.2% cheaper |
| End-to-end time (lower is better) | 62.29s | 6.65s | −89.3% Mercury 2 is 89.3% 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 (Low): 64.1% less intelligence, 91.2% less cost to run the AA benchmark, and 89.3% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Kimi K3 (Low) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- 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 (Low)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-low/mercury-2 · Markdown: /api/v1/compare/kimi-k3-low/mercury-2?format=md · llms.txt · agent skill