GPT-6 Luna (Non-reasoning) vs Hy3
Which is better for the money: GPT-6 Luna (Non-reasoning) by OpenAI 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 6.8 points higher (25.3 vs 18.5) but costs 7.8× as much to run the benchmark ($150 vs $19.2; mid-range vs budget cost classes). They are not alternatives at one budget: pick Hy3 when you need intelligence above 18.5, and GPT-6 Luna (Non-reasoning) when 18.5 is enough.
| Metric | GPT-6 Luna (Non-reasoning) | Hy3 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 18.5 | 25.3 | +36.8% Hy3 is 36.8% more intelligent |
| Total AA cost (lower is better) | $19.16 | $149.61 | +680.8% Hy3 is 7.8× more expensive |
| End-to-end time (lower is better) | 5.02s | 33.73s | +571.9% Hy3 is 571.9% slower |
| Price per 1M tokens (in / out) | $0.10 / $0.50 | $0.14 / $0.56 | |
| Input modalities | Text, Image | Text |
Hy3 vs GPT-6 Luna (Non-reasoning): 36.8% more intelligence, 680.8% more cost to run the AA benchmark, and 571.9% slower end-to-end.
Today's Pareto frontier
Against the frontier
- GPT-6 Luna (Non-reasoning) is off the frontier: at the same or lower cost, GPT-6 Luna (Low) scores 21.50 for $10.6.
- Hy3 is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
GPT-6 Luna (Non-reasoning)Hy3Today'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/gpt-6-luna-non-reasoning/hy3 · Markdown: /api/v1/compare/gpt-6-luna-non-reasoning/hy3?format=md · llms.txt · agent skill