Hy3 vs Mistral Small 4 (Reasoning)
Which is better for the money: Hy3 by Tencent or Mistral Small 4 (Reasoning) by Mistral? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-24.
Trade-off
Verdict
Depends on your budget
Hy3 scores 14.0 points higher (25.3 vs 11.3) but costs 3.5× as much to run the benchmark ($150 vs $42.5; mid-range vs budget cost classes). They are not alternatives at one budget: pick Hy3 when you need intelligence above 11.3, and Mistral Small 4 (Reasoning) when 11.3 is enough.
| Metric | Hy3 | Mistral Small 4 (Reasoning) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 25.3 | 11.3 | −55.3% Mistral Small 4 (Reas… is 55.3% less intelligent |
| Total AA cost (lower is better) | $149.61 | $42.50 | −71.6% Mistral Small 4 (Reas… is 71.6% cheaper |
| End-to-end time (lower is better) | 33.29s | 14.50s | −56.4% Mistral Small 4 (Reas… is 56.4% faster |
| Price per 1M tokens (in / out) | $0.14 / $0.56 | $0.15 / $0.60 | |
| Input modalities | Text | Text, Image |
Mistral Small 4 (Reasoning) vs Hy3: 55.3% less intelligence, 71.6% less cost to run the AA benchmark, and 56.4% faster end-to-end.
Today's Pareto frontier
Against the frontier
- Hy3 is off the frontier: at the same or lower cost, GPT-6 Luna (Max) scores 38.10 for $122.
- Mistral Small 4 (Reasoning) is off the frontier: at the same or lower cost, GPT-6 Luna (Medium) scores 29.90 for $31.2.
Hy3Mistral Small 4 (Reasoning)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/hy3/mistral-small-4 · Markdown: /api/v1/compare/hy3/mistral-small-4?format=md · llms.txt · agent skill