DeepSeek V4 Flash 0731 vs DeepSeek V4 Pro 0813
Select two models to compare their coding performance, pricing, and features.
Key Takeaways
| Metric | DeepSeek V4 Flash 0731 | DeepSeek V4 Pro 0813 |
|---|---|---|
| Coding Index | 100.00 | 100.69 |
| Coding tier | B+ | B+ |
| Input Price | $0.44/1M | $1.32/1M |
| Output Price | $1.32/1M | $3.96/1M |
| Context Window | 1.31M | 1.05M |
Coding Index
Coding Index · Reference baseline: 100
Integer index difference: 0.69
DeepSeek V4 Flash 0731 = 100. The index reflects two shared tests, not capability multiples or statistically established gaps.
Verdict
Both models share an editorial tier. The evidence does not establish a reliable order within it; compare them on your own projects.
DeepSeek V4 Flash 0731
Off-peak: $0.22 / $0.66 · 1M
Peak: Monday–Friday, UTC 01:00–04:00 and 06:00–10:00; off-peak at other times. Input prices assume a cache miss.
Benchmark evidence and ranking method · DeepSeek V4 Flash 0731
Cross-checked
Artificial Analysis · DeepSeek V4 Flash 0731 (Reasoning, Max Effort)
Test configuration: max
Terminal-Bench 2.1: 78.65% · SciCode: 50.35%
LiveBench · 2026-06-25 · deepseek-v4-flash-0731
Coding: 74.98% · Agentic Coding: 46.77%
ModelRank assigns editorial tiers using public coding evidence, not automatic index thresholds or statistically proven classes. Price, speed, context size and sponsorship do not affect tiers.
DeepSeek V4 Flash 0731 = 100. The index reflects two shared tests, not capability multiples or statistically established gaps.
Results reflect particular model configurations. Agent harnesses, reasoning effort, tools, budgets and tasks may differ. Close indices do not establish superiority. B describes a relative position in this selected group, not an inability to code.
Data checked: 2026-09-10. Snapshot date is the retrieval date, not necessarily the evaluation date.
DeepSeek V4 Pro 0813
Off-peak: $0.66 / $1.98 · 1M
Peak: Monday–Friday, UTC 01:00–04:00 and 06:00–10:00; off-peak at other times. Input prices assume a cache miss.
Benchmark evidence and ranking method · DeepSeek V4 Pro 0813
Cross-checked
Artificial Analysis · DeepSeek V4 Pro 0813 (Reasoning, Max Effort)
Test configuration: max
Terminal-Bench 2.1: 78.65% · SciCode: 51.04%
LiveBench · 2026-06-25 · deepseek-v4-pro-0813
Coding: 77.16% · Agentic Coding: 54.95%
ModelRank assigns editorial tiers using public coding evidence, not automatic index thresholds or statistically proven classes. Price, speed, context size and sponsorship do not affect tiers.
DeepSeek V4 Flash 0731 = 100. The index reflects two shared tests, not capability multiples or statistically established gaps.
Results reflect particular model configurations. Agent harnesses, reasoning effort, tools, budgets and tasks may differ. Close indices do not establish superiority. B describes a relative position in this selected group, not an inability to code.
Data checked: 2026-09-10. Snapshot date is the retrieval date, not necessarily the evaluation date.
Results reflect particular model configurations. Agent harnesses, reasoning effort, tools, budgets and tasks may differ. Close indices do not establish superiority. B describes a relative position in this selected group, not an inability to code.
API cost estimate
1M input + 200K output (USD):
DeepSeek V4 Flash 0731: $0.7040
DeepSeek V4 Pro 0813: $2.1120
Estimates use the displayed channel and time band; DeepSeek uses peak rates. Cache discounts, tool fees, long-context price tiers and uncounted additional tokens are excluded.
Frequently Asked Questions
Which model is better for coding: DeepSeek V4 Flash 0731 or DeepSeek V4 Pro 0813?
DeepSeek V4 Pro 0813 has a higher reference index on the selected shared benchmark configuration than DeepSeek V4 Flash 0731 (100.69 vs 100.00, a 0.69-point difference). This does not by itself establish general coding superiority. Editorial tiers: DeepSeek V4 Flash 0731 = B+; DeepSeek V4 Pro 0813 = B+. Tiers are separate editorial assessments and do not establish a reliable order within a tier.
How do DeepSeek V4 Flash 0731 and DeepSeek V4 Pro 0813 compare in API pricing and costs?
DeepSeek V4 Flash 0731: $0.44/1M input, $1.32/1M output (Official · peak); DeepSeek V4 Pro 0813: $1.32/1M input, $3.96/1M output (Official · peak). 1M input + 200K output (USD): $0.7040 vs $2.1120.
What are the context window sizes of DeepSeek V4 Flash 0731 and DeepSeek V4 Pro 0813?
DeepSeek V4 Flash 0731 provides a 1,310,720 token context window, while DeepSeek V4 Pro 0813 provides 1,048,576 tokens.