Kimi K2.7 Code vs Kimi K3 (max)
Which is better for the money: Kimi K2.7 Code by Kimi or Kimi K3 (max) by Kimi? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-09-09.
Trade-off
Verdict
Depends on your budget
Kimi K3 (max) scores 17.5 points higher (43.8 vs 26.3) but costs 3.7× as much to run the benchmark ($3658 vs $993; flagship vs premium cost classes). They are not alternatives at one budget: pick Kimi K3 (max) when you need intelligence above 26.3, and Kimi K2.7 Code when 26.3 is enough.
| Metric | Kimi K2.7 Code | Kimi K3 (max) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 26.3 | 43.8 | +66.5% Kimi K3 (max) is 66.5% more intelligent |
| Total AA cost (lower is better) | $992.88 | $3,658.07 | +268.4% Kimi K3 (max) is 3.7× more expensive |
| End-to-end time (lower is better) | 41.22s | 63.46s | +54.0% Kimi K3 (max) is 54.0% slower |
| Price per 1M tokens (in / out) | $0.95 / $4.00 | $3.00 / $15.0 | |
| Input modalities | Not reported | Not reported |
Kimi K3 (max) vs Kimi K2.7 Code: 66.5% more intelligence, 268.4% more cost to run the AA benchmark, and 54.0% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Kimi K2.7 Code is off the frontier: at the same or lower cost, GLM-5.3-Flash scores 41.90 for $280.
- Kimi K3 (max) is off the frontier: at the same or lower cost, GPT-6 Astra (high) scores 51.00 for $2917.
Kimi K2.7 CodeKimi K3 (max)Today'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-k2-7-code/kimi-k3 · Markdown: /api/v1/compare/kimi-k2-7-code/kimi-k3?format=md · llms.txt · agent skill