$060 min
- Up to 1M tokens
- The clock starts at your first API call
- One trial per person, no payment needed
Free trial60 min or 1M tokens →
DeepSeek & GLM API
DeepSeek & GLM API
at 2% of list price.
One OpenAI- and Anthropic-compatible key for DeepSeek V4.1 Flash and GLM-5.3 Flash. Prepaid in USDT or USDC, per token or flat-rate with Unlimited at $10 for 30 days. No card, no auto-renewal.
New accounts get a free 1-hour trial with up to 1M tokens; the clock starts at your first API call.
Independent resellerNot affiliated with DeepSeek or Z.ai.
req 7f3a·e1deepseek-v4.1-flash
uncached input48,210 tok
cached input31,744 tok
output2,906 tok
list price$0.018141
charged$0.000363
Pricing
List price × 0.02
We take the vendor's published per-token price and charge 2% of it. No platform fee, no markup on cached tokens, no rounding up to the cent.
2%of list price
- Billed per token to the millionth of a cent
- Prepaid, $5 to $2000 per top-up
- Balance never expires
from $10/ 30 days
- No per-token charge while it is active
- 2 to 100 slots: 2 for $10, 5 for $20, 10 for $40, +$4 per extra slot
- Paid up front, no auto-renewal
| Model | Input | Cached input | Output |
|---|---|---|---|
| DeepSeek V4.1 Flashdeepseek-v4.1-flash | $0.006$0.3 | $0.00012$0.006 | $0.024$1.2 |
| GLM-5.3 Flashglm-5.3-flash | $0.003$0.15 | $0.0006$0.03 | $0.01$0.5 |
USD per 1M tokens; the struck price is the vendor's list price. Full pricing and limits
Paying with coin
First time paying in crypto?
You send a stablecoin from a wallet or an exchange to the address on the checkout page. Four steps, about five minutes, no card and no ID.
01
Choose an amount
Top up $5 to $2000 from the dashboard, or buy Unlimited.
02
Pick coin and network
Only these are watched:
USDT · TronUSDC · Solana
03
Send the exact amount
Copy the address and the amount to the last digit and send exactly the amount shown before the timer runs out. A different amount is not credited automatically; it is held for manual review, not lost.
04
Credited automatically
The page updates by itself once the network confirms the transfer. Usually a few minutes.
Wrong network means lost funds
The same coin exists on several networks. Send only on the network your order shows; a transfer on another network goes to an address we do not control there.
USDT · TronUSDC · SolanaUSDT · EthereumUSDC · BaseUSDC · Ethereum
Exchange fees break the match
The last digits of the amount identify your payment. If an exchange subtracts its fee, the amount that arrives no longer matches. Add the fee or send from a wallet.
5.004570example
Quickstart
Keep your code. Change two lines.
Anything that speaks the OpenAI or Anthropic API works: the official SDKs, Claude Code, OpenCode, Cursor, LangChain, plain curl.
From sign-up to first response
- Create an account with an email. No phone number.
- Create a key in the dashboard.
- Set the base URL to
https://unlimited-deepseek.com/v1and use one ofdeepseek-v4.1-flashorglm-5.3-flash.
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"}]}'
# 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="deepseek-v4.1-flash",
messages=[{"role": "user", "content": "Hello"}],
)
print(r.choices[0].message.content)
# pip install anthropic
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": "Hello"}],
)
FAQ
Before you pay
Why is it this cheap?
Our cost to serve these models is far below the public list price, and we pass most of that on. The price is pegged to the vendor's list price: when they cut, we cut.
Is there a free trial?
New accounts get a free 1-hour trial with up to 1M tokens; the clock starts at your first API call. The trial ends at whichever comes first, the time limit or the token cap, and one trial is granted per person. After that you pay per token from a prepaid balance.
Do you keep my prompts?
No. We store request metadata needed for billing — model, token counts, cost, latency, status — and never the prompt or completion text.
Which coins and networks?
USDT · Tron or USDC · Solana. The only networks watched for this site are Tron or Solana; stablecoins there have low network fees.
What if I send the wrong amount?
The exact amount shown is what identifies your payment, so send it to the cent and digit. An amount that differs is not credited automatically: it is kept for manual review, not lost. Keep the transaction id: it identifies your transfer.
What is the Unlimited plan?
Unlimited is a 30-day plan paid up front in USDT or USDC, priced by how many requests 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. Request limits still apply. Your prepaid balance works as before.
What happens when a plan ends?
Nothing renews automatically, and we never charge you again by ourselves. When the 30 days are over, requests go back to per-token billing from your prepaid balance. To keep unlimited use, pay for another period before the end and it is added after your current end date. 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. The dashboard shows a notice when the plan is close to ending.
What are the request limits?
On every plan a request may carry up to 200,000 input tokens; longer input is rejected with HTTP 400 that states the limit. Output is capped at 64,000 tokens: a larger max_tokens is lowered to that cap, not rejected. The per-key rate limit applies as well.
What does slot limit mean?
A slot is one request in progress. Unlimited comes with 2 to 100 slots; you choose the count when you pay. A request that arrives while all your slots are busy is rejected at once with HTTP 429 and the message "slot limit"; it is not queued. Retry when one of your running requests has finished.
Policies
What you can check
Status
Live check of the API, the model upstream and billing, run each time you open the status page.
Open status →Prompts are not stored
We keep model, token counts, cost, latency and status for billing. Never the prompt or completion text.
Pegged to list price
Every price is 2% of the vendor's published rate, checked 2026-11-05.
See pricing →