How this DeepSeek Token Calculator works
DeepSeek V4.1 Flash and V4 Pro price tiers
DeepSeek's current text API presents a compact two-model choice. V4.1 Flash is the throughput-oriented tier at off-peak cache-miss $0.15 input and $0.6 output; peak cache-miss $0.3 input and $1.2 output per 1M tokens; V4 Pro is published at off-peak cache-miss $0.66 input and $1.98 output; peak cache-miss $1.32 input and $3.96 output per 1M tokens for harder reasoning, coding, and long-horizon work. The central store defines both UTC peak windows and the weekday rule, and the calculator derives its default peak share from that weekly schedule. Both endpoints support thinking and non-thinking modes, so the bill is shaped less by a separate reasoning SKU than by which tier you select and how many completion tokens the task consumes.
Why DeepSeek output matters even at low rates
DeepSeek bills output separately from cache-miss input, and the ratio differs by model. A retrieval workflow with a large document and a short answer can remain inexpensive, while an agent that reasons verbosely, rewrites files, or produces several drafts shifts more of its spend to output. The calculator separates the two meters so a shorter completion limit can be tested without pretending the prompt shrank. Thinking tokens reported as output also belong in the output estimate when measuring real calls.
Cache hits change the DeepSeek bill
DeepSeek automatically discounts input tokens that match reusable context in its disk cache. The store retains both cache-hit and cache-miss input rates for off-peak and peak periods; this calculator deliberately estimates with cache-miss input, giving a conservative baseline for a new or changing prompt. Repeated system instructions, reference material, and conversation prefixes can make the invoice lower, but the hit rate depends on measured traffic and should not be guessed.
When DeepSeek is cheap—and when Pro can be expensive
V4.1 Flash is the active low-cost choice for high-volume processing, especially when outputs are compact and repeated context earns cache hits. V4 Flash is retired; its legacy API name temporarily routes to V4.1 Flash. V4 Pro is the higher-priced option inside DeepSeek's lineup. Test quality on your own data, route routine requests to V4.1 Flash, and promote only the cases that benefit from Pro. Tool fees, retries, regional taxes, and any third-party gateway markup remain outside this token-only estimate.