DeepSeek V4-Pro: DeepSeek's Paid API With Peak and Off-Peak Rates
Key facts · verified Oct 29, 2026
DeepSeek V4-Pro is a DeepSeek model released on 13 Aug 2026. Its API price is $1.32 per million input tokens and $3.96 per million output tokens (peak, cache miss; $0.66 / $1.98 off-peak). Its context window is 1M tokens. It ranks #65 on LMArena's text leaderboard (Elo 1458, read 3 Nov 2026). Source: api-docs.deepseek.com.
DeepSeek's V4-Pro, a paid API model with a 1M token context window and 384k max output. Reached general availability on 13 August 2026 as deepseek-v4-pro, and moved to peak/off-peak billing on 16 August: $1.32 input / $3.96 output per 1M tokens at peak, halved to $0.66 / $1.98 off-peak.
Key Features
What is DeepSeek V4-Pro?
DeepSeek V4-Pro is a model from the Chinese AI lab DeepSeek, offered as a paid API with a 1 million token context window. Under peak/off-peak billing introduced on 16 August 2026, standard rates are $1.32 per million input tokens and $3.96 per million output tokens, halving automatically to $0.66 / $1.98 outside peak hours. DeepSeek's documentation names no open-source licence and publishes no weights for it.
Cached Input Pricing
At peak hours DeepSeek charges $1.32 per million input tokens on a cache miss and $0.044 per million on a cache hit - about 30x less. Off-peak every rate halves: $0.66 cache-miss input and $0.022 cached. Output is $3.96 per million tokens at peak, $1.98 off-peak.
What DeepSeek Does Not Publish
DeepSeek's pricing page publishes no coding or math benchmark scores for V4-Pro, no architecture details and no multimodal capability detail, so this guide makes no claim on any of them.
Context Window
V4-Pro offers a 1 million token context window, the same as Claude Opus 5; GPT-5.6 lists 1.05 million.
Ranking and Price Against Other Models
On LMArena's text leaderboard DeepSeek V4-Pro has 1458 Elo (rank 65); Claude Opus 5 has 1492 Elo (rank 14, high effort), Gemini 3.1 Pro 1487 Elo (rank 20) and GPT-5.6 Sol 1485 Elo (rank 22, xhigh effort). Published API prices per million input / output tokens: V4-Pro $1.32 / $3.96 at peak and $0.66 / $1.98 off-peak, Gemini 3.1 Pro $2 / $12, Claude Opus 5 $5 / $25.
Key Takeaways
- A paid API model - DeepSeek publishes no open-source licence or weights for V4-Pro
- Priced at $1.32 / $3.96 per 1M input / output tokens at peak; $0.66 / $1.98 off-peak
- Cached input drops to $0.044 per 1M tokens at peak ($0.022 off-peak)
- LMArena: 1458 Elo (rank 65)
- 1M token context window
Official Resources
Full Specifications
$1.32 / $3.96 per million tokens | |
|---|---|
| Identity | |
| Developer | DeepSeek |
| Released | Aug 2026 |
| Status | GA |
| Licence | MIT |
| Self-hostable | Yes - open weights on Hugging Face |
| Cost | |
| Blended $/1M tokensinput × 0.75 + output × 0.25 | $1.980 / 1M tokens |
| Input price | $1.320 / 1M tokens |
| Output price | $3.960 / 1M tokens |
| Cached input | $0.044 / 1M tokens (peak); $0.022 off-peak |
| Batch discount | Off-peak billing: 50% off all rates outside 01:00-04:00 and 06:00-10:00 UTC, Monday to Friday (weekends and Chinese public holidays are off-peak in full) |
| Free tier | Not offered |
| Capacity | |
| Context window | 1M tokens |
| Max output | 384k tokens |
| Long-context surcharge | None (flat rate across the 1M window) |
| Capability | |
| Vision in | No |
| Audio in | No |
| Function calling | Yes |
| Structured output | Yes |
| Extended reasoning | Yes - thinking effort levels low, high and max |
| Web search | No |
| Code execution | No |
| Access | |
| API | Yes |
| Chat app | Yes |
| Cloud marketplaces | Not offered |
| Fine-tuning | Not offered |
| Measured quality | |
| LMArena Elo | 1458 (checked Nov 2026) |
| LMArena rank | Rank 65 |
| GPQA Diamond | 91 |
| SWE-bench Verified | 78 · max effort |
| Elo per dollarLMArena Elo ÷ blended $/1M tokens | 736 |