DeepSeek V4 Flash 0731 (Max) vs DeepSeek V4 Pro 0813 (Non-reasoning)
Which is better for the money: DeepSeek V4 Flash 0731 (Max) by DeepSeek or DeepSeek V4 Pro 0813 (Non-reasoning) by DeepSeek? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Clear winner
Verdict
Pick DeepSeek V4 Flash 0731 (Max)
DeepSeek V4 Pro 0813 (Non-reasoning) vs DeepSeek V4 Flash 0731 (Max): 40.5% less intelligence, 125.8% more cost to run the AA benchmark, and 61.8% faster end-to-end.
| Metric | DeepSeek V4 Flash 0731 (Max) | DeepSeek V4 Pro 0813 (Non-reasoning) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 34.3 | 20.4 | −40.5% DeepSeek V4 Pro 0813 … is 40.5% less intelligent |
| Total AA cost (lower is better) | $199.63 | $450.76 | +125.8% DeepSeek V4 Pro 0813 … is 2.3× more expensive |
| End-to-end time (lower is better) | 17.94s | 6.85s | −61.8% DeepSeek V4 Pro 0813 … is 61.8% faster |
| Price per 1M tokens (in / out) | $0.14 / $0.28 | $1.32 / $3.96 | |
| Input modalities | Text | Text |
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.
- DeepSeek V4 Pro 0813 (Non-reasoning) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Medium) scores 47.80 for $361.
DeepSeek V4 Flash 0731 (Max)DeepSeek V4 Pro 0813 (Non-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/deepseek-v4-flash/deepseek-v4-pro-non-reasoning · Markdown: /api/v1/compare/deepseek-v4-flash/deepseek-v4-pro-non-reasoning?format=md · llms.txt · agent skill