Celeris-1 vs Hy3
Which is better for the money: Celeris-1 by Celeris or Hy3 by Tencent? 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
Hy3 scores 19.0 points higher (25.3 vs 6.3) but costs 2.5× as much to run the benchmark ($150 vs $60.0; both mid-range class). Pick Hy3 when the extra intelligence is worth the premium, otherwise Celeris-1.
| Metric | Celeris-1 | Hy3 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 6.3 | 25.3 | +301.6% Hy3 is 4.0× more intelligent |
| Total AA cost (lower is better) | $59.99 | $149.61 | +149.4% Hy3 is 2.5× more expensive |
| End-to-end time (lower is better) | 0.87s | 33.73s | +3777.0% Hy3 is 3777.0% slower |
| Price per 1M tokens (in / out) | $0.20 / $0.70 | $0.14 / $0.56 | |
| Input modalities | Text, Image | Text |
Hy3 vs Celeris-1: 301.6% more intelligence, 149.4% more cost to run the AA benchmark, and 3777.0% 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.
- Hy3 is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
Celeris-1Hy3Today'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/hy3 · Markdown: /api/v1/compare/celeris-1/hy3?format=md · llms.txt · agent skill