Text (LLM)$1.06 / $3.16per 1M
DeepSeek V4 Pro API
DeepSeek V4 Pro is DeepSeek's larger text model for work that needs deeper reasoning. 1M-token context, text input, with function calling and JSON output.
- Input → Output
- In / out price
- $1.06 / $3.16per 1M
- Context
- 1M
- Max output
- 384K
Supported parameters
Values this model accepts through the API.
| Parameter | Accepted values |
|---|---|
| messages | Text · up to 1M tokens |
| stream | true to receive the answer as a stream (SSE) |
| max_tokens | Caps output including reasoning, up to 384K tokens |
| thinking | { "type": "disabled" } turns reasoning off (on by default) |
| tools | OpenAI-style function calling |
| response_format | json_object or json_schema |
| temperature | Controls how varied the answer is |
Pricing
$1.06 / $3.16per 1M
| Item | Price per 1M tokens |
|---|---|
| Input | $1.06 |
| Cached input | $0.036 |
| Output | $3.16 |
Billed on the tokens actually used. Reasoning tokens count as output. Each request adds its exact cost to a running total, and whole credits are taken from your balance when that total reaches 1 credit; the remainder carries over. Failed requests are not charged.
One Fuzion credit balance covers the web app, the API, and MCP. No separate top-up · 1 credit = ฿1
Use it through the API
Same format as OpenAI Chat Completions. Point the base URL at Fuzion and use your API key, and the OpenAI SDK works as is.
curl -X POST https://fuzionhub.co/api/v1/chat/completions \
-H "Authorization: Bearer $FUZION_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "byteplus/deepseek-v4-pro",
"messages": [
{
"role": "user",
"content": "Explain what an API is in two sentences."
}
]
}'from openai import OpenAI
client = OpenAI(base_url="https://fuzionhub.co/api/v1", api_key="<FUZION_API_KEY>")
stream = client.chat.completions.create(
model="byteplus/deepseek-v4-pro",
messages=[{"role": "user", "content": "Explain what an API is in two sentences."}],
stream=True,
)
for chunk in stream:
if chunk.choices and chunk.choices[0].delta.content:
print(chunk.choices[0].delta.content, end="")Frequently asked questions
What is DeepSeek V4 Pro?
DeepSeek V4 Pro is DeepSeek's larger text model for work that needs deeper reasoning. 1M-token context, text input, with function calling and JSON output.
How much does DeepSeek V4 Pro cost?
Input $1.06 and output $3.16 per 1M tokens, billed on the tokens actually used. 1 Fuzion credit = ฿1, and the same balance pays for video, image, and audio.
How long can a prompt for DeepSeek V4 Pro be?
The context window is 1M tokens and one answer can be up to 384K tokens, reasoning included.
What inputs does DeepSeek V4 Pro accept?
Text only.
How do I use DeepSeek V4 Pro through the API?
Create an API key in your Fuzion account, then call POST /api/v1/chat/completions with the model "byteplus/deepseek-v4-pro". The format matches OpenAI, so the OpenAI SDK works once you change the base URL. Streaming, function calling, and JSON output are supported.
How are credits charged?
A single request costs a fraction of a credit, so costs are added up and whole credits are taken when the running total reaches 1 credit. The remainder carries over. Failed requests are not charged.