DeepSeek V4 Pro 0813 API Pricing & Monthly Cost Simulator
Verified 2026 unit economics for DeepSeek DeepSeek V4 Pro 0813 (deepseek/deepseek-v4-pro-0813): $0.111/1M uncached input, $0.097/1M cached prefix read, and $4.44/1M output across a 1,048,576-token context window (released 2026-08-12).
Uncached Input (/1M)
$0.111
Cached Prompt Read (/1M)
$0.097 (-13%)
Output Completion (/1M)
$4.44
50% Batch API (In / Out)
Real-time API only (no 50% async batch)
Interactive 30-Day Monthly Spend Simulator
Live Unit-Cost & Prompt Cache Slider (DeepSeek V4 Pro 0813)
deepseek/deepseek-v4-pro-0813
Estimated 30-day spend
$119.04/mo
Cost per 1,000 calls
$3.9680
Monthly savings vs 0% cache
$0.84/mo
Technical Limits & API Capabilities
Context window
1,048,576
Max output tokens
943,718
Workload tier
Frontier Reasoning
Input modalities
text
Tool Calling: Yes
JSON Schema Strict: Yes
Reasoning Tokens: Supported
Batch API: Unavailable
Same-Category 2026 Model Alternatives (Frontier Reasoning)
| Model | Provider | Input / Cached ($/1M) | Output ($/1M) | Context / Max Out |
|---|---|---|---|---|
| DeepSeek V4 Pro 0813 | DeepSeek | $0.111 / $0.097 | $4.44 | 1,048,576 / 943,718 |
| Claude Fable 5.2 | Anthropic | $8 / $0.2 | $40 | 1,000,000 / 128,000 |
| GLM 5.3 FlashX Legal | Z.ai / GLM | $0.37 / $0.075 | $1.25 | 1,048,576 / 131,072 |
| GPT-6.1 Terra Pro | OpenAI | $1 / $0.05 | $5 | 1,050,000 / 128,000 |
| Kimi K3.1 Pro | Moonshot | $2 / $0.2 | $10 | 1,048,576 / 943,718 |
| GPT-6.1 Sol Pro | OpenAI | $2 / $0.1 | $10 | 1,050,000 / 128,000 |
| Claude Sonnet 5.5 | Anthropic | $2 / $0.1 | $10 | 1,000,000 / 128,000 |
Head-to-Head Matchups Featuring DeepSeek V4 Pro 0813
- DeepSeek V4 Pro 0813 vs DeepSeek V4.1 Flash — $0.111 vs $0.3 / 1M
- Kimi K3.1 Pro vs DeepSeek V4 Pro 0813 — $2 vs $0.111 / 1M
Frequently Asked Questions
How much does DeepSeek V4 Pro 0813 cost per 1 million tokens in 2026?
DeepSeek V4 Pro 0813 by DeepSeek is billed at $0.111 per 1M uncached input tokens, $0.097 per 1M cached input read tokens (13% cache discount), and $4.44 per 1M output tokens.
What are the context window and max output token limits for DeepSeek V4 Pro 0813?
DeepSeek V4 Pro 0813 supports a 1,048,576-token input context window and up to 943,718 completion output tokens per request, accepting text inputs.
Does DeepSeek V4 Pro 0813 support strict JSON Schema outputs, tool calling, and Batch API?
DeepSeek V4 Pro 0813 capability profile: tool calling = Supported, strict JSON Schema structured outputs = Supported, reasoning mode = Supported, and 50% asynchronous Batch API = Not listed.