DeepSeek V4 Pro API Pricing: Low Cost with V4.1 Flash and Hy4 Flash
On Onysoft, DeepSeek V4 Pro costs $0.99 per 1 million input tokens and $2.97 per 1 million output tokens. DeepSeek V4.1 Flash is $0.225 / $0.90 and Tencent Hy4 Flash is $0.0525 / $0.2108. All three offer a 1M-token context window and work with a single Onysoft API key on an OpenAI-compatible endpoint.
This post covers the prices of the three models as of November 20, 2026, DeepSeek's peak-hour pricing, a sample monthly cost and how to send your first request. Current prices are always in the model catalog.
At a Glance: Price Table
| Model | Onysoft ID | Input (1M tokens) | Output (1M tokens) | Context |
|---|---|---|---|---|
| DeepSeek V4 Pro | deepseek/deepseek-v4-pro-0813 | $0.99 | $2.97 | 1M |
| DeepSeek V4.1 Flash | deepseek/deepseek-v4.1-flash | $0.225 | $0.90 | 1M |
| Tencent Hy4 Flash | tencent/hy4-flash | $0.0525 | $0.2108 | 1M |
Prices are Onysoft user prices as of November 20, 2026, in US dollars. The DeepSeek rows show the off-peak price; see the next section. The Hy4 Flash price is valid until November 30, 2026.
Usage is deducted from your Onysoft balance.
DeepSeek Peak-Hour Pricing
The price of DeepSeek models changes during the day. The table above shows the off-peak price. On weekdays between 04:00–07:00 and 09:00–13:00 (Türkiye time, UTC+3) peak-hour pricing applies and the price doubles: $1.98 / $5.94 for DeepSeek V4 Pro and $0.45 / $1.80 for DeepSeek V4.1 Flash.
The price can also vary with the infrastructure that serves the request. The model catalog shows a range for these models; the upper end of the range is a ceiling that is never exceeded. Moving batch jobs and scheduled tasks outside peak hours lowers cost directly.
Hy4 Flash: Introductory Price Until November 30
Tencent Hy4 Flash has the lowest per-token price of the three: $0.0525 per 1 million input tokens and $0.2108 per 1 million output tokens. It has a 1M-token context window and a 64K-token maximum output.
This is an introductory price valid until November 30, 2026. After that date the price becomes $0.105 per 1 million input tokens and $0.4215 per 1 million output tokens. Onysoft is a Tencent Cloud partner and serves Hy4 Flash through a direct integration with Tencent Cloud; see the partnership post.
Sample Monthly Cost
For an application that uses 10 million input tokens and 2 million output tokens a month (DeepSeek rows at the off-peak price):
| Model | Input | Output | Monthly total |
|---|---|---|---|
| DeepSeek V4 Pro | 10 × 0.99 = $9.90 | 2 × 2.97 = $5.94 | $15.84 |
| DeepSeek V4.1 Flash | 10 × 0.225 = $2.25 | 2 × 0.90 = $1.80 | $4.05 |
| Tencent Hy4 Flash | 10 × 0.0525 = $0.53 | 2 × 0.2108 = $0.42 | $0.95 |
Use the cost calculator for your own volume. To compare quality and cost across the three models with the same prompt, change only the model field of the request.
How to Use It: First Request
Create an account, generate an API key with the sk-ony- prefix in the dashboard and add balance. A curl request to DeepSeek V4.1 Flash:
curl https://api.onysoft.com/v1/chat/completions \
-H "Authorization: Bearer sk-ony-YOUR_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "deepseek/deepseek-v4.1-flash",
"max_tokens": 4096,
"messages": [{"role": "user", "content": "Summarize this customer email in three bullet points: ..."}]
}'A request to DeepSeek V4 Pro with the official openai Python package:
from openai import OpenAI
client = OpenAI(
base_url="https://api.onysoft.com/v1",
api_key="sk-ony-YOUR_KEY",
)
response = client.chat.completions.create(
model="deepseek/deepseek-v4-pro-0813",
messages=[{"role": "user", "content": "Write an error-handling checklist for a REST API."}],
)
print(response.choices[0].message.content)For Hy4 Flash, set the model name to tencent/hy4-flash. Parameters and error codes are in the API documentation; to try without writing code, use the Playground.
Frequently Asked Questions
How much does the DeepSeek V4 Pro API cost?
On Onysoft, DeepSeek V4 Pro (deepseek/deepseek-v4-pro-0813) costs $0.99 per 1 million input tokens and $2.97 per 1 million output tokens as of November 20, 2026. On weekdays between 04:00–07:00 and 09:00–13:00 (UTC+3) peak-hour pricing doubles the price.
How much does the DeepSeek V4.1 Flash API cost?
On Onysoft, DeepSeek V4.1 Flash (deepseek/deepseek-v4.1-flash) costs $0.225 per 1 million input tokens and $0.90 per 1 million output tokens. During peak hours the price is $0.45 / $1.80.
What is the Hy4 Flash price and how long is it valid?
Tencent Hy4 Flash (tencent/hy4-flash) costs $0.0525 per 1 million input tokens and $0.2108 per 1 million output tokens. This introductory price is valid until November 30, 2026; after that the price is $0.105 / $0.4215.
What is the context window of these models?
DeepSeek V4 Pro, DeepSeek V4.1 Flash and Tencent Hy4 Flash all offer a 1M-token context window.
Does my existing OpenAI-compatible code work with these models?
Yes. Set base_url to https://api.onysoft.com/v1, api_key to your Onysoft key and model to deepseek/deepseek-v4-pro-0813, deepseek/deepseek-v4.1-flash or tencent/hy4-flash.
Share this article
Related pages
Ready to build?
Access 820+ AI models through a single API. Pay as you go — no subscription.