Quickstart

Base URL https://unlimited-deepseek.com/v1. Authenticate with Authorization: Bearer cp-… or x-api-key: cp-…. Create a key in the dashboard.

New accounts get a free 1-hour trial with up to 1M tokens; the clock starts at your first API call.

curl

curl https://unlimited-deepseek.com/v1/chat/completions \
  -H "Authorization: Bearer $API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model": "deepseek-v4.1-flash", "messages": [{"role": "user", "content": "Hello"}]}'

OpenAI SDK

# pip install openai
from openai import OpenAI

client = OpenAI(base_url="https://unlimited-deepseek.com/v1", api_key="cp-...")
r = client.chat.completions.create(
    model="glm-5.3-flash",
    messages=[{"role": "user", "content": "Write a haiku about ledgers"}],
    stream=True,
)
for chunk in r:
    print(chunk.choices[0].delta.content or "", end="")

// npm i openai
import OpenAI from "openai";
const client = new OpenAI({ baseURL: "https://unlimited-deepseek.com/v1", apiKey: process.env.API_KEY });
const r = await client.chat.completions.create({ model: "deepseek-v4.1-flash", messages: [{ role: "user", content: "Hi" }] });

Anthropic SDK

import anthropic
client = anthropic.Anthropic(base_url="https://unlimited-deepseek.com", api_key="cp-...")
msg = client.messages.create(model="deepseek-v4.1-flash", max_tokens=1024,
                             messages=[{"role": "user", "content": "Hi"}])

Claude Code

export ANTHROPIC_BASE_URL=https://unlimited-deepseek.com
export ANTHROPIC_AUTH_TOKEN=cp-...
export ANTHROPIC_MODEL=deepseek-v4.1-flash
export ANTHROPIC_SMALL_FAST_MODEL=glm-5.3-flash
claude

OpenCode

Add a provider to ~/.config/opencode/opencode.json:

{
  "provider": {
    "unlimited-deepseek": {
      "npm": "@ai-sdk/openai-compatible",
      "options": { "baseURL": "https://unlimited-deepseek.com/v1", "apiKey": "{env:API_KEY}" },
      "models": { "deepseek-v4.1-flash": {}, "glm-5.3-flash": {} }
    }
  }
}

Models

Endpoints: POST /v1/chat/completions, POST /v1/messages, GET /v1/models. Streaming responses include a final usage chunk, which is what you are billed for.

Limits and plans

Every plan, pay as you go and the free trial included, has the same request limits: input of at most 200,000 tokens per request and output of at most 64,000 tokens. A longer input is rejected with HTTP 400 input_too_long. A max_tokens above 64,000 is not an error: it is lowered to 64,000.

Unlimited runs 30 days, paid up front in stablecoins, with no auto-renewal. It is priced by slots, the number of requests that may run at the same time: 2 slots $10, 5 slots $20, 10 slots $40, then $4 per extra slot up to 100. While it is active there is no per-token charge. A request that would need another slot is rejected immediately with HTTP 429 slot_limit; it is never queued. When the plan ends, requests are billed per token from your balance again. Changing the slot count while a plan is active starts the new plan at once for 30 days, and the days you had not used yet are converted into extra days at the two plans' prices (12 unused days of a $10 plan become 3 extra days on a $40 plan). Paying for the same slot count again adds 30 days after your current end. Subscribe in the dashboard.

{"error": {"code": "input_too_long", "message": "..."}}   # 400
{"error": {"code": "slot_limit", "message": "slot limit ..."}}   # 429

Errors