deepseek-v4-pro API
deepseek-v4-pro is DeepSeek's flagship reasoning model with a 1M-token context window. On CaMeL Hub you call it through the OpenAI-compatible API at https://api.camel-hub.com/v1 for $3 per 1M input tokens and $6 per 1M output tokens — one API key, no separate DeepSeek account required.
Pricing
| per 1M input tokens | per 1M output tokens | |
|---|---|---|
| CaMeL Hub | $3.00 | $6.00 |
| Provider list price | $0.435 | $0.87 |
Prices are the public default-group rate in USD per 1M tokens; input and output are billed separately. · Provider list price · verified 2026-07-11
Specifications
- Provider
- DeepSeek
- Context window
- 1000000 tokens
- Max output
- 384000 tokens
- Capabilities
- streaming, reasoning
- Endpoints
- openai
What it is good for
- Reviewing an entire repository in one request — load hundreds of source files into the 1M-token context and ask for cross-file bug analysis or a refactoring plan
- Multi-step math, logic, and algorithm problems where thinking mode's deliberate reasoning beats fast pattern-matching answers
- Contract, filing, or research-corpus analysis: feed book-length documents whole instead of chunking them through a retrieval pipeline
- Producing very long structured deliverables — full technical specs, audit reports, or complete documentation — within the up-to-384K-token output budget
- The thinking stage of agent pipelines: let v4-pro reason through a plan or verdict, then hand execution to lighter models or your own code
Quickstart
OpenAI-compatible: set base_url to https://api.camel-hub.com/v1 and use your CaMeL Hub API key. No SDK changes needed.
cURL
curl https://api.camel-hub.com/v1/chat/completions \
-H "Authorization: Bearer $CAMEL_HUB_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek-v4-pro",
"messages": [{"role": "user", "content": "Hello!"}]
}'Python
from openai import OpenAI
client = OpenAI(
api_key="$CAMEL_HUB_KEY",
base_url="https://api.camel-hub.com/v1", # <-- the only change vs. the OpenAI default
)
resp = client.chat.completions.create(
model="deepseek-v4-pro",
messages=[{"role": "user", "content": "Hello!"}],
)
print(resp.choices[0].message.content)Node.js
import OpenAI from "openai";
const client = new OpenAI({
apiKey: process.env.CAMEL_HUB_KEY,
baseURL: "https://api.camel-hub.com/v1", // <-- the only change vs. the OpenAI default
});
const resp = await client.chat.completions.create({
model: "deepseek-v4-pro",
messages: [{ role: "user", content: "Hello!" }],
});
console.log(resp.choices[0].message.content);Get an API key Browse integration guides
Frequently asked questions
How is deepseek-v4-pro billed on CaMeL Hub?
Input and output tokens are metered separately: $3 per 1M input tokens and $6 per 1M output tokens. Everything the model returns — including reasoning it emits before the final answer — counts as output. The usage field in each response reports the exact token counts you were charged for.
Does deepseek-v4-pro support streaming and reasoning?
Yes to both. Pass stream: true for token-by-token streaming — recommended, because a reasoning model can think for a while before the final answer starts. Deep reasoning (thinking mode) is the model's core capability and works through the standard chat completions API; no special endpoint is needed.
What are the context window and output limits?
DeepSeek's documentation lists a 1M-token context length and a maximum of 384K output tokens for deepseek-v4-pro. In practice that means book-length inputs and report-length outputs are possible in a single call.
How do I start using deepseek-v4-pro?
Create an account at https://api.camel-hub.com/console, generate an API key, then point any OpenAI-compatible SDK at base_url https://api.camel-hub.com/v1 with model "deepseek-v4-pro". Existing OpenAI client code needs no other changes. Top-ups support Alipay, WeChat Pay, and international cards.
When should I pick deepseek-v4-pro over deepseek-v4-flash?
Pick v4-pro when reasoning depth is the bottleneck: competition-grade math, subtle code defects, high-stakes analysis. deepseek-v4-flash is the speed-oriented tier for everyday, high-volume tasks. Both are served through the same CaMeL Hub endpoint, so switching is a one-line model-name change.