DeepSeek · 1M context · deep thinking mode · OpenAI-compatible · pay-as-you-go
DeepSeek official list price vs. Waterseller relay price. Verified Nov 2026 · official pricing page.
| Direction | Official | Waterseller | You save |
|---|---|---|---|
| Input | $0.15 | $0.0375 | 75% off |
| Output | $0.6 | $0.15 | 75% off |
Cache-aware billing supported where the vendor offers it. 1M context · deep thinking mode.
DeepSeek's off-peak pricing tier made available around the clock. Exceptional price-to-intelligence for reasoning-heavy workloads.
One line changes: point base_url at Waterseller. Everything else is stock OpenAI SDK.
from openai import OpenAI
client = OpenAI(
api_key="sk-...", # your Waterseller key
base_url="https://waterseller.cc/v1" # one-line switch
)
resp = client.chat.completions.create(
model="deepseek-v4.1-flash",
messages=[{"role": "user", "content": "Hello!"}]
)
print(resp.choices[0].message.content)
Common questions about DeepSeek V4.1 Flash API relay pricing.
Input: $0.0375 per million tokens. Output: $0.15 per million tokens. That's 75% below DeepSeek's official list price of $0.15 / $0.6. Pay-as-you-go, no minimum spend, no subscription.
Register a Waterseller account, create a key on the Tokens page, and point your OpenAI SDK base_url to https://waterseller.cc/v1. New accounts get $1 in free credit — enough to test before paying anything.
Yes — identical weights and identical official endpoints. We operate a relay that bills at wholesale rates; nothing about the model, its context window, or its capabilities is modified.
All official features are supported: streaming (SSE), function/tool calling, JSON mode, and vision where the vendor supports it. Latency matches direct access.
Live in 5 minutes · one-line switch from OpenAI SDK
Get your API key → No credit card required