DeepSeek V4 API access to DeepSeek's open-source model family — V4 Pro, V4 Flash and V4.1 Flash chat models plus the R1 reasoning model, delivering frontier-level performance at a fraction of the cost.
What's on your mind?
Ask APIMart
deepseek-v4.1-flash
Transparent pricing with no hidden fees. Pay only for what you use.
Your timezone: UTC
Peak
Mon–Sun · 00:00–14:00
Standard time rate · 20% total savings now
Off-peak
All other times
Extra 50% off
Per 1M tokens
| Rate | Peak | Off-peak | Official Price |
|---|---|---|---|
| Input | 2.2857CreditsPer 1M tokens ~$0.2286 | 1.1429CreditsPer 1M tokens ~$0.1143 | 2.8571CreditsPer 1M tokens ~$0.2857 |
| Cached input | 0.2286CreditsPer 1M tokens ~$0.0229 | 0.1143CreditsPer 1M tokens ~$0.0114 | 0.2857CreditsPer 1M tokens ~$0.0286 |
| Output | 9.1429CreditsPer 1M tokens ~$0.9143 | 4.5714CreditsPer 1M tokens ~$0.4571 | 11.4286CreditsPer 1M tokens ~$1.1429 |
Access DeepSeek V4 Pro for complex tasks, DeepSeek V4.1 Flash for fast, low-cost chat, and DeepSeek-R1 for deep reasoning. Open-source models at a fraction of the cost.
50K+
Active Users
99.9%
Uptime
2x
Faster
70%
Cost Savings
Why DeepSeek is the top open-source choice for AI developers
How teams use DeepSeek models in production
Start using DeepSeek models in minutes
Create your free APIMart account and top up your balance.
Generate an API key to access DeepSeek models programmatically.
Use the playground above or integrate via the OpenAI-compatible API.
What developers say about our DeepSeek API
“DeepSeek-R1 matches o1 on our math benchmarks at 1/10 the cost. Incredible value.”
Alex Chen
ML Engineer
“V3 is our default model for general chat. Fast, capable, and extremely affordable.”
Sarah Li
Product Manager
“DeepSeek V4.1 Flash is perfect for our research assistant product. Fast answers and strong long-context analysis.”
Mike Wang
Senior Developer
“We run R1-distill-32b for high-volume tasks. Same great reasoning at a lower price point.”
Emily Zhou
AI Engineer
“DeepSeek's coding ability is surprisingly strong. It handles complex refactoring tasks well.”
David Liu
Tech Lead
“Best price-to-performance ratio in the market. We migrated our entire pipeline to DeepSeek.”
Lisa Zhang
CTO
Common questions about using the DeepSeek API
We offer the latest DeepSeek V4.1 Flash, V4 Flash, and V4 Pro, plus DeepSeek-R1 (reasoning), DeepSeek-V3.2, and DeepSeek-OCR. Earlier versions such as V3 and V3.1 remain available.
R1 is DeepSeek's reasoning model that uses chain-of-thought to solve complex math, logic, and coding problems step by step.
Token-based pricing (input + output). DeepSeek models are among the most cost-effective frontier models available.
Yes. DeepSeek models use the standard chat completions format and work with the OpenAI SDK.
Yes. Your API data is not stored or used for training. All requests are encrypted.
DeepSeek V4.1 Flash or V4 Flash for fast, low-cost chat. DeepSeek V4 Pro for complex tasks. DeepSeek-R1 for step-by-step reasoning. DeepSeek-OCR for text recognition in images and documents.
In APIMart's model catalog, the DeepSeek 4 generation offers long context: V4 Pro and V4.1 Flash accept up to 1M input tokens, and V4 Flash accepts up to 128K. Choose V4 Pro or V4.1 Flash for long documents and large codebases, and check the pricing table above for each model's rates.
Yes. The DeepSeek API is billed per million tokens, with input and output tokens priced separately. The pricing table on this page lists every model and specification, and the prices shown are Gold member prices (20% off). Platinum and Diamond members save even more, up to 28%, and you only pay for successful requests.
Sign up for an APIMart account, open the API Keys page and create a key. The same key works for the DeepSeek API and every other model on APIMart, with pay-as-you-go billing and no subscription.
DeepSeek models on APIMart use the OpenAI-compatible Chat Completions format. Use your APIMart key, pick a DeepSeek model such as DeepSeek V4 Pro or DeepSeek V4.1 Flash, and send your messages with the OpenAI SDK or any HTTP client. Model IDs, parameters, and code examples are in the APIMart docs (docs.apimart.ai).
Explore more models in the same category.
Claude Haiku 5.5
Claude Haiku 5.5 API on APIMart delivers fast, cost-efficient AI, offering responsive conversations, capable coding assistance, and reliable performance for high-volume workflows.
Kimi K3
Kimi-K3 is a next-generation large language model launched by Moonshot AI. It features ultra-long context, multimodal understanding, and strong coding capabilities, making it suitable for complex reasoning, software development, knowledge analysis, and agent automation tasks.
Claude Opus 5
Claude Opus 5 is Anthropic’s next-generation flagship large language model, featuring enhanced capabilities in code development, complex reasoning, knowledge analysis, and agent execution, making it ideal for large-scale project development and professional work scenarios.
Gemini 3.5 Flash Lite
A lightweight multimodal model launched by Google that emphasizes low cost, low latency, and high throughput. It is suitable for document parsing, data extraction, structured output, and large-scale agent workflows, delivering faster response times while maintaining core reasoning capabilities.