DeepSeek V4 Pro API
Frontier reasoning model. Best for agents, long-context analysis and code generation.
$0.660 per 1M input and $1.980 per 1M output. 1M context. 50% below DeepSeek's published rate.
Call it in three lines
from openai import OpenAI
client = OpenAI(
base_url="https://api.tokenify.dev/v1",
api_key=os.environ["TOKENIFY_API_KEY"],
)
response = client.chat.completions.create(
model="deepseek/deepseek-v4-pro",
messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)What it costs
A typical agent turn — a 4,000-token prompt with 2,500 of it cached, producing 600 tokens — costs $0.00229. A million of those turns costs $2288.
- Input / 1M
- $1.320$0.660
- Output / 1M
- $3.960$1.980
- Cache read / 1M
- $0.044
50% offoff DeepSeek’s published rate on input and output. Cache reads pass through undiscounted — they already cost a fraction of a fresh token.
How it compares
| Model | Input | Output | Cache read | Context |
|---|---|---|---|---|
| DeepSeek V4.1 Flash | $0.150 | $0.600 | $0.006 | 1M |
| DeepSeek V4 Pro | $0.660 | $1.980 | $0.044 | 1M |
| DeepSeek V4 Flash | $0.220 | $0.660 | $0.014 | 1M |
Peak rates, per 1M tokens, from the live catalogue. All models · the full rate card · how sellers of this model compare.
Questions about this model
Is this the same model as DeepSeek's own API?
Yes — DeepSeek V4 Pro is DeepSeek's model, served through our endpoint rather than theirs. We do not fine-tune, quantise or substitute it. What differs is the price, the API formats on offer and how it is billed.
How much cheaper is this than DeepSeek's list price?
50% on input and output, blended three parts prompt to one part completion. Cache reads are not discounted: they are billed at DeepSeek's published cache rate, which is already a small fraction of a fresh token.
Can I use it with Claude Code?
Yes. DeepSeek V4 Pro answers on the Anthropic Messages API as well as the OpenAI one, so Claude Code reaches it with three environment variables and no router. The Claude Code page has the setup and the compatibility results.
Do I need a subscription?
No. Billing is per token from prepaid credit, with no subscription, no seats and no minimum. Credit does not expire.
Last updated 2026-10-01. Prices come from the live catalogue at the time this page was built.