Deep-thinking reasoning tier with 98% prefix-cache discount and 50% diurnal off-peak pricing
deepseek-reasoner| Scenario | Input Tokens | Output Tokens | Uncached Cost | With Prompt Caching |
|---|---|---|---|---|
| Short Chat Query | 1,000 | 500 | $0.00049 | $0.00024 |
| Document Summarization | 10,000 | 2,000 | $0.00364 | $0.00112 |
| Codebase & Context Analysis | 100,000 | 20,000 | $0.0364 | $0.0112 |
| Batch Corpus Processing | 1,000,000 | 100,000 | $0.3220 | $0.0700 |
Tier 1 (free tier) reference limits — your account's actual quota may be higher.
Ideal for production workloads demanding reasoning capabilities, deep context depth (256k tokens), and reliability from DeepSeek. Excellent when predictable tokenomics and prompt caching support are paramount.
If your use-case requires sub-second streaming latency or ultra-high frequency classification at micro-cent pricing, consider lighter budget options such as Gemini Flash-Lite or Claude Haiku. For deep formal logic, consider dedicated reasoning models like o3.
For DeepSeek-V4.1-Reasoner, 1 million input tokens costs $0.28, while 1 million output tokens costs $0.42. If using prompt caching, repetitive input prefixes are discounted to $0.028 per million.
DeepSeek-V4.1-Reasoner features a maximum context window of 262,144 tokens (~196,608 words), with a maximum output limit of 384,000 tokens per completion.
DeepSeek-V4.1-Reasoner utilizes the DeepSeek Byte-Level BPE (102k vocabulary). Token counting on TokenMath runs client-side to ensure maximum privacy.
Yes. DeepSeek-V4.1-Reasoner supports prompt caching with a cached input rate of $0.028/1M (saving up to 90% on repeated input context).