DeepSeek V4.1 Flash (Non-reasoning) vs GPT-5.4 mini (Xhigh)
Which is better for the money: DeepSeek V4.1 Flash (Non-reasoning) by DeepSeek or GPT-5.4 mini (Xhigh) by OpenAI? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-25.
Clear winner
Verdict
Pick DeepSeek V4.1 Flash (Non-reasoning)
GPT-5.4 mini (Xhigh) vs DeepSeek V4.1 Flash (Non-reasoning): 2.4% less intelligence, 537.9% more cost to run the AA benchmark, and 3877.7% slower end-to-end.
| Metric | DeepSeek V4.1 Flash (Non-reasoning) | GPT-5.4 mini (Xhigh) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 24.7 | 24.1 | −2.4% GPT-5.4 mini (Xhigh) is 2.4% less intelligent |
| Total AA cost (lower is better) | $238.91 | $1,524.02 | +537.9% GPT-5.4 mini (Xhigh) is 6.4× more expensive |
| End-to-end time (lower is better) | 2.74s | 108.99s | +3877.7% GPT-5.4 mini (Xhigh) is 3877.7% slower |
| Price per 1M tokens (in / out) | $0.30 / $1.20 | $0.75 / $4.50 | |
| Input modalities | Text, Image | Text, Image |
Today's Pareto frontier
Against the frontier
- DeepSeek V4.1 Flash (Non-reasoning) is off the frontier: at the same or lower cost, MiMo-V2.6-Pro scores 46.30 for $207.
- GPT-5.4 mini (Xhigh) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
DeepSeek V4.1 Flash (Non-reasoning)GPT-5.4 mini (Xhigh)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-v4-1-flash-non-reasoning/gpt-5-4-mini · Markdown: /api/v1/compare/deepseek-v4-1-flash-non-reasoning/gpt-5-4-mini?format=md · llms.txt · agent skill