DeepSeek API — call DeepSeek V4.1 Flash and other models with one key
cocodot now calls DeepSeek V4.1 Flash. Use the OpenAI-compatible API with base_url https://cocodot.co/api/ai/v1 and model deepseek-v4.1-flash. The same key and balance also call Claude, GPT, Gemini, Qwen, GLM and more — switching models means changing the model field.
One key, several model families
The same cocodot key and balance call DeepSeek, Claude, GPT, Gemini, Qwen and GLM. Switching models means changing one field — base_url and the rest of your code stay the same.
Which model name to use
Set model to deepseek-v4.1-flash, or simply deepseek. Only V4.1 Flash is open right now: naming another DeepSeek version returns HTTP 400, is not charged, and is never quietly switched to a different model.
How billing works
Billed by actual token usage; per-model rates are on the pricing page. Top up with Alipay / WeChat Pay (RMB) or USDT / USDC.
Call it in one line
OpenAI-compatible — set the model name, keep the rest.
from openai import OpenAI
client = OpenAI(
base_url="https://cocodot.co/api/ai/v1",
api_key="YOUR_COCODOT_KEY",
)
resp = client.chat.completions.create(
model="deepseek-v4.1-flash", # or simply "deepseek"
messages=[{"role": "user", "content": "hi"}],
)
print(resp.choices[0].message.content)FAQ
Can I call DeepSeek on cocodot?
Yes. DeepSeek V4.1 Flash is open now — set model to deepseek-v4.1-flash (or deepseek).
How do I connect?
It is an OpenAI-compatible API: set base_url to https://cocodot.co/api/ai/v1, api_key to your cocodot key and model to deepseek-v4.1-flash; everything else follows the OpenAI SDK.
Are other DeepSeek versions supported?
Only V4.1 Flash is open right now. Naming another version (for example one with a date suffix or a newer version number) returns HTTP 400, is not charged and is never switched to a different model. We will update this page when a new version opens.