deepseek-ai/DeepSeek-V3
deepseek-ai/DeepSeek-V3 is a DeepSeek chat and completion model you can call through flatkey.ai's OpenAI-compatible API. On flatkey it runs at $0.133333 per 1M input tokens — about 47% below the $0.25 official list price, and $0.533333 per 1M output tokens. One API key covers GPT, Claude, Gemini, DeepSeek, Seedance and 500+ other models, so you can switch models without changing SDKs or juggling separate accounts. Billing is prepaid with live usage analytics and a single invoice, which keeps spend predictable as you scale.
Pricing
Specifications
- Provider
- DeepSeek
- Modality
- Text / chat
- Access
- OpenAI-compatible API
- Endpoints
- openai
How to use the deepseek-ai/DeepSeek-V3 API
- 1Create a flatkey.ai account and API keySign up for free, then generate an API key in the console. New accounts get bonus credit on their first top-up.
- 2Point your OpenAI client at flatkeySet base_url to https://router.flatkey.ai/v1 and use your flatkey key as the API key. No other code changes are needed — the API is OpenAI-compatible.
- 3Call deepseek-ai/DeepSeek-V3Send a request with model set to "deepseek-ai/DeepSeek-V3" using the OpenAI SDK or curl. Copy a ready-to-run example below.
curl "https://router.flatkey.ai/v1/chat/completions" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $FLATKEY_API_KEY" \
-d '{"model":"deepseek-ai/DeepSeek-V3","messages":[{"role":"user","content":"Say hello in one sentence."}]}'deepseek-ai/DeepSeek-V3 vs similar models
| Model | Input / 1M |
|---|---|
| deepseek-ai/DeepSeek-V3 | $0.133333 |
| deepseek-v3.2-thinking | $0.154667 |
| deepseek-ocr | $40 |
| deepseek-v3.2-exp | $40 |
| deepseek-v3.2 | $0.141867 |
Availability (last 30 days)
Frequently asked questions
How much does the deepseek-ai/DeepSeek-V3 API cost?
On flatkey.ai, deepseek-ai/DeepSeek-V3 costs $0.133333 per 1M input tokens and $0.533333 per 1M output tokens, roughly 47% below the official list price. Top up $200 and get $100 free to stretch the budget further.
Is deepseek-ai/DeepSeek-V3 OpenAI-compatible?
Yes. Keep your existing OpenAI SDK, switch base_url to https://router.flatkey.ai/v1, and use your flatkey key. The model id stays "deepseek-ai/DeepSeek-V3".
Is deepseek-ai/DeepSeek-V3 free to use?
Usage is pay-as-you-go, but priced well below the official list (about 47% cheaper), and new accounts receive bonus credit on their first top-up, so you can start testing at low cost.
How do I get an API key for deepseek-ai/DeepSeek-V3?
Sign up on flatkey.ai, create an API key in the console, and add prepaid balance. You can start calling deepseek-ai/DeepSeek-V3 in minutes.
What other models can I use with the same key?
One flatkey key gives you GPT, Claude, Gemini, DeepSeek, Seedance and 500+ other models through a single OpenAI-compatible endpoint and one invoice.
How reliable is deepseek-ai/DeepSeek-V3 on flatkey?
We publish the live 30-day success rate, throughput and latency for deepseek-ai/DeepSeek-V3 directly on this page, updated from real traffic.
More DeepSeek models
Start building with deepseek-ai/DeepSeek-V3
One OpenAI-compatible key, prepaid billing, and bonus credit on your first top-up. Sign up in under a minute.
Get your API key