Celeris-1 vs Gemini 3.1 Flash-Lite
Which is better for the money: Celeris-1 by Celeris or Gemini 3.1 Flash-Lite by Google? 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
Gemini 3.1 Flash-Lite scores 9.3 points higher (15.6 vs 6.3) but costs 2.6× as much to run the benchmark ($156 vs $60.0; both mid-range class). Pick Gemini 3.1 Flash-Lite when the extra intelligence is worth the premium, otherwise Celeris-1.
| Metric | Celeris-1 | Gemini 3.1 Flash-Lite | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 15.6 | +147.6% Gemini 3.1 Flash-Lite is 2.5× more intelligent |
| Total AA cost (lower is better) | $59.99 | $155.87 | +159.8% Gemini 3.1 Flash-Lite is 2.6× more expensive |
| End-to-end time (lower is better) | 0.87s | 6.06s | +596.6% Gemini 3.1 Flash-Lite is 596.6% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $0.25 / $1.50 | |
| Input modalities | Text, Image | Text, Image, Speech, Video |
Gemini 3.1 Flash-Lite vs Celeris-1: 147.6% more intelligence, 159.8% more cost to run the AA benchmark, and 596.6% 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.
- Gemini 3.1 Flash-Lite is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
Celeris-1Gemini 3.1 Flash-LiteToday's Pareto 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/gemini-3-1-flash-lite-preview · Markdown: /api/v1/compare/celeris-1/gemini-3-1-flash-lite-preview?format=md · llms.txt · agent skill