DeepSeek V4 Flash 0731 (Max) vs Qwen3.8 Max
Which is better for the money: DeepSeek V4 Flash 0731 (Max) by DeepSeek or Qwen3.8 Max by Alibaba? 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
Qwen3.8 Max scores 5.9 points higher (40.2 vs 34.3) but costs 16× as much to run the benchmark ($3172 vs $200; flagship vs mid-range cost classes). They are not alternatives at one budget: pick Qwen3.8 Max when you need intelligence above 34.3, and DeepSeek V4 Flash 0731 (Max) when 34.3 is enough.
| Metric | DeepSeek V4 Flash 0731 (Max) | Qwen3.8 Max | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 34.3 | 40.2 | +17.2% Qwen3.8 Max is 17.2% more intelligent |
| Total AA cost (lower is better) | $199.63 | $3,172.37 | +1489.1% Qwen3.8 Max is 16× more expensive |
| End-to-end time (lower is better) | 17.94s | 64.64s | +260.3% Qwen3.8 Max is 260.3% slower |
| Price per 1M tokens (in / out) | $0.14 / $0.28 | $2.00 / $6.00 | |
| Input modalities | Text | Text, Image, Video |
Qwen3.8 Max vs DeepSeek V4 Flash 0731 (Max): 17.2% more intelligence, 1489.1% more cost to run the AA benchmark, and 260.3% slower end-to-end.
Today's Pareto frontier
Against the frontier
- DeepSeek V4 Flash 0731 (Max) is off the frontier: at the same or lower cost, Claude Haiku 5.5 (Xhigh) scores 41.20 for $158.
- Qwen3.8 Max is off the frontier: at the same or lower cost, Kimi K3.1 Pro scores 55.10 for $3117.
DeepSeek V4 Flash 0731 (Max)Qwen3.8 MaxToday'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-flash/qwen3-8-max-0803 · Markdown: /api/v1/compare/deepseek-v4-flash/qwen3-8-max-0803?format=md · llms.txt · agent skill