Claude 4.5 Sonnet (Reasoning) vs DeepSeek V4 Pro 0813 (Non-reasoning)
Which is better for the money: Claude 4.5 Sonnet (Reasoning) by Anthropic or DeepSeek V4 Pro 0813 (Non-reasoning) by DeepSeek? Intelligence, total benchmark cost, and end-to-end time from Artificial Analysis & other sources, snapshot 2026-10-26.
Depends on your budget
Claude 4.5 Sonnet (Reasoning) scores 0.3 points higher (20.7 vs 20.4) but costs 3.3× as much to run the benchmark ($1501 vs $451; premium vs mid-range cost classes). They are not alternatives at one budget: pick Claude 4.5 Sonnet (Reasoning) when you need intelligence above 20.4, and DeepSeek V4 Pro 0813 (Non-reasoning) when 20.4 is enough.
| Metric | Claude 4.5 Sonnet (Reasoning) | DeepSeek V4 Pro 0813 (Non-reasoning) | Difference |
|---|---|---|---|
| Intelligence (higher is better) | 20.7 | 20.4 | −1.4% DeepSeek V4 Pro 0813 … is 1.4% less intelligent |
| Total AA cost (lower is better) | $1,501.22 | $450.76 | −70.0% DeepSeek V4 Pro 0813 … is 70.0% cheaper |
| End-to-end time (lower is better) | 21.00s | 6.74s | −67.9% DeepSeek V4 Pro 0813 … is 67.9% faster |
| Price per 1M tokens (in / out) | $3.00 / $15.0 | $1.32 / $3.96 | |
| Input modalities | Text, Image | Text |
DeepSeek V4 Pro 0813 (Non-reasoning) vs Claude 4.5 Sonnet (Reasoning): 1.4% less intelligence, 70.0% less cost to run the AA benchmark, and 67.9% faster end-to-end.
Against the frontier
- Claude 4.5 Sonnet (Reasoning) is off the frontier: at the same or lower cost, GPT-6.1 Sol (Max) scores 51.80 for $1082.
- 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.
Compare something else
Accept: text/markdown or ?format=md returns this page as Markdown. JSON: /api/v1/compare/claude-4-5-sonnet-thinking/deepseek-v4-pro-non-reasoning · Markdown: /api/v1/compare/claude-4-5-sonnet-thinking/deepseek-v4-pro-non-reasoning?format=md · llms.txt · agent skill