Celeris-1 vs Kimi K3 (Low)
Which is better for the money: Celeris-1 by Celeris or Kimi K3 (Low) by Kimi? 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 23.8 points higher (30.1 vs 6.3) but costs 20× as much to run the benchmark ($1190 vs $60.0; premium vs mid-range cost classes). They are not alternatives at one budget: pick Kimi K3 (Low) when you need intelligence above 6.3, and Celeris-1 when 6.3 is enough.
| Metric | Celeris-1 | Kimi K3 (Low) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 30.1 | +377.8% Kimi K3 (Low) is 4.8× more intelligent |
| Total AA cost (lower is better) | $59.99 | $1,190.42 | +1884.4% Kimi K3 (Low) is 20× more expensive |
| End-to-end time (lower is better) | 0.87s | 62.29s | +7059.8% Kimi K3 (Low) is 7059.8% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $3.00 / $15.0 | |
| Input modalities | Text, Image | Text, Image |
Kimi K3 (Low) vs Celeris-1: 377.8% more intelligence, 1884.4% more cost to run the AA benchmark, and 7059.8% slower end-to-end.
Today's Pareto frontier
Against the frontier
- Celeris-1 is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Medium) scores 34.50 for $56.4.
- Kimi K3 (Low) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
Celeris-1Kimi K3 (Low)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/celeris-1/kimi-k3-low · Markdown: /api/v1/compare/celeris-1/kimi-k3-low?format=md · llms.txt · agent skill