Deepseek V4 Pro API Pricing & Quick Start

Name: Deepseek V4 Pro API
Brand: DeepSeek

DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...

Intelligence#49 / 638

Artificial Analysis Intelligence Index

Speed#200 / 329

72.3

Output tokens per second (median)

Input Price#320 / 437

$2.40NT$ 76

USD / 1M tokens

Output Price#284 / 437

$4.80NT$ 152

USD / 1M tokens

First-token latency#116 / 329

1.77s

Time to first token (AA median)

Provider	DeepSeek
Released	April 2026
Model ID	`deepseek/deepseek-v4-pro`

Technical specifications

Context window

1M tokens

Reasoning

Yes

Input

Text

Output

Text

Pricing comparison

BazaarLink retail price alongside AA's observed market standard.

BazaarLink price our site

Input$2.40NT$ 76

Output$4.80NT$ 152

Cache read$0.14NT$ 4

Cache write—

3:1 blended (est.)$3.00NT$ 95

All prices per 1M tokens. Supports TWD credit card / official invoice.

Market reference Artificial Analysis

Input$1.32

Output$3.96

Cache readAA not tracked

Cache writeAA not tracked

3:1 blended—

Our price differs from AA reference by ~82%.
Source: artificialanalysis.ai

Independent benchmarks

Scores below come from Artificial Analysis, an independent third party. BazaarLink does not participate in the testing.

Intelligence 36

Coding 69

Math

MMLU

GPQA

This modelCategory leader

Coding ability

#54out of 255 models

LiveCodeBench / SciCode and similar

Overall intelligence

#49out of 638 models

Aggregated across academic benchmarks

Time to first token

#116out of 329 models

AA 全球 p50：1.77 秒

Intelligence

Top 8%

This model

Category leader

Median

Coding

Top 21%

This model

Category leader

Median

Top 5 — CodingCoding Index leaderboard

1Claude Fable 5.1 (Adaptive Reasoning, Max Effort, Default Fallback)?/claude-fable-5-182$10 / $50

2Claude Fable 5.1 (Adaptive Reasoning, Xhigh Effort, Default Fallback)?/claude-fable-5-1-xhigh81$10 / $50

3Claude Fable 5.1 (Adaptive Reasoning, High Effort, Default Fallback)?/claude-fable-5-1-high79$10 / $50

4GPT-5.6 Sol (xhigh)openai/gpt-5-6-sol-xhigh78$4 / $20

5Claude Opus 5 (Adaptive Reasoning, Max Effort)anthropic/claude-opus-578$5 / $25

AASource: Artificial Analysis · Independent third-party evaluation; not influenced by BazaarLinkView full evaluation on AA →

Quick Start

Use Deepseek V4 Pro via BazaarLink API — just change the base_url:

Python

from openai import OpenAI

client = OpenAI(
    base_url="https://api.bazaarlink.ai/v1",
    api_key="sk-bl-YOUR_API_KEY",
)

response = client.chat.completions.create(
    model="deepseek/deepseek-v4-pro",
    messages=[{"role": "user", "content": "Hello!"}],
)
print(response.choices[0].message.content)

TypeScript

import OpenAI from "openai";

const client = new OpenAI({
  baseURL: "https://api.bazaarlink.ai/v1",
  apiKey: "sk-bl-YOUR_API_KEY",
});

const response = await client.chat.completions.create({
  model: "deepseek/deepseek-v4-pro",
  messages: [{ role: "user", content: "Hello!" }],
});
console.log(response.choices[0].message.content);

Why use Deepseek V4 Pro via BazaarLink?

✓USD billing (TWD-quoted) + unified invoices — no foreign credit card needed for Taiwan teams
✓OpenAI-compatible API — zero code changes required
✓Automatic failover — multi-provider redundancy for the same model
✓Chinese-language support — local team, instant help

Try Now ← All Models

Frequently Asked Questions

What is Deepseek V4 Pro?

How much does the Deepseek V4 Pro API cost?

Deepseek V4 Pro costs $2.40 per 1M input tokens and $4.80 per 1M output tokens when accessed through BazaarLink.

How do I use Deepseek V4 Pro with the OpenAI SDK?

Set base_url to "https://api.bazaarlink.ai/v1" and use model ID "deepseek/deepseek-v4-pro". All OpenAI SDK methods (chat.completions, embeddings, streaming) work without code changes.

What is the context window for Deepseek V4 Pro?

Deepseek V4 Pro supports a context window of 1,048,576 tokens.

Is Deepseek V4 Pro available for free?

Deepseek V4 Pro is a paid model. BazaarLink offers free trial credits on registration so you can test it without a credit card.

Relay integrity checks

Tested endpoints: 852 runs
Independent hosts: 345
Model-family fingerprint match: 70%
Anomalies detected: 630 runs
Data range: past 90 days

BazaarLink Probe verifies endpoints that claim to serve this model and flags behavior that differs from the reference baseline.

View the full integrity report →