DeepSeek V4 Pro 0424 (Max) vs Hy3
Which is better for the money: DeepSeek V4 Pro 0424 (Max) by DeepSeek 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
DeepSeek V4 Pro 0424 (Max) scores 5.1 points higher (30.4 vs 25.3) but costs 1.9× as much to run the benchmark ($277 vs $150; both mid-range class). Pick DeepSeek V4 Pro 0424 (Max) when the extra intelligence is worth the premium, otherwise Hy3.
| Metric | DeepSeek V4 Pro 0424 (Max) | Hy3 | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 30.4 | 25.3 | −16.8% Hy3 is 16.8% less intelligent |
| Total AA cost (lower is better) | $277.30 | $149.61 | −46.0% Hy3 is 46.0% cheaper |
| End-to-end time (lower is better) | 48.90s | 33.13s | −32.2% Hy3 is 32.2% faster |
| Price per 1M tokens (in / out) | $0.43 / $0.87 | $0.14 / $0.56 | |
| Input modalities | Text | Text |
Hy3 vs DeepSeek V4 Pro 0424 (Max): 16.8% less intelligence, 46.0% less cost to run the AA benchmark, and 32.2% faster end-to-end.
Today's Pareto frontier
Against the frontier
- DeepSeek V4 Pro 0424 (Max) is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
- Hy3 is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
DeepSeek V4 Pro 0424 (Max)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/deepseek-v4-pro-0424/hy3 · Markdown: /api/v1/compare/deepseek-v4-pro-0424/hy3?format=md · llms.txt · agent skill