Native DeepSeek Web 11 models live Unlimited tokens Instant delivery
DeepSeek Infrastructure · Production ready

The DeepSeek layer
for your stack

Native DeepSeek Web API. Unlimited tokens. OpenAI compatible.
11 models online → · Recommended: Cline

11
Models
Tokens
24/7
Uptime
SSE
Streaming

DeepSeek Models

Live list from API — only models that work right now.

deepseek-chat

Fast chat — код, боты, IDE, повседневные запросы.

Files

deepseek-chat-search

Web search enabled — актуальная информация из интернета.

SearchFiles

deepseek-default

Fast chat — код, боты, IDE, повседневные запросы.

Files

deepseek-default-search

Web search enabled — актуальная информация из интернета.

SearchFiles

deepseek-expert

Expert mode — максимальное качество на тяжёлых задачах.

Chat

deepseek-r1

Reasoning mode — complex logic, math, multi-step analysis.

ThinkingFiles

deepseek-r1-search

Reasoning mode — complex logic, math, multi-step analysis.

ThinkingSearchFiles

deepseek-reasoner

Reasoning mode — complex logic, math, multi-step analysis.

ThinkingFiles

deepseek-reasoner-search

Reasoning mode — complex logic, math, multi-step analysis.

ThinkingSearchFiles

deepseek-v3

Fast chat — код, боты, IDE, повседневные запросы.

Files

deepseek-v4-pro

Reasoning mode — complex logic, math, multi-step analysis.

Thinking

GET /v1/model-capabilities · 11 supported

Works with your stack

Cline · recommendedCursorPythonTelegramOpen WebUIVS Code
Platform

Built for scale

Premium DeepSeek access — clean Web API, no routing surprises.

Unlimited tokens

No token billing. One key — full access for your session.

Drop-in compatible

OpenAI Chat Completions — change base URL and API key.

Secure keys

HTTPS, isolated keys dllm-…

API endpoint

Production proxy. Streaming SSE supported.

deepseek.llm-api.fun/v1/chat/completions
POST /v1/chat/completions
Authorization: Bearer dllm-•••
{
  "model": "deepseek-chat",
  "messages": [{ "role": "user", "content": "..." }]
}

11 models

Chat, reasoning, expert, search — pick the right model.

Documentation

Integration guides

Same settings everywhere. Cline is our top pick from client feedback.

URL: https://deepseek.llm-api.fun/v1
Model: deepseek-chat
Key: dllm-…
Recommended by clients

Cline + deepseek-chat — stable tool calls for coding agents.

Cline

  1. Provider: OpenAI Compatible
  2. Base URL: https://deepseek.llm-api.fun/v1
  3. Model: deepseek-chat
  4. API Key: dllm-…

Cursor

  1. Override OpenAI Base URL: https://deepseek.llm-api.fun/v1
  2. Model: deepseek-chat
  3. Key: dllm-…

Python

test.py
import requests

API_KEY = "dllm-XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX"

r = requests.post(
    "https://deepseek.llm-api.fun/v1/chat/completions",
    headers={"Authorization": f"Bearer {API_KEY}", "Content-Type": "application/json"},
    json={"model": "deepseek-chat", "messages": [{"role": "user", "content": "hello"}]},
    timeout=120,
)
print(r.status_code, r.json())

Telegram / bots

  1. URL: https://deepseek.llm-api.fun/v1/chat/completions
  2. Method: POST
  3. Model: deepseek-chat

Get API access

Instant delivery · key dllm-…

Need agent routing?

For long projects with token balance — use the main LLM API.

Support dllm- key required
Enter your dllm- key and describe the issue.