DeepSeek V4 Pro (0813) API in India

DeepSeek V4 Pro (0813) is the flagship of DeepSeek's V4 family, and it is available on unoblox as deepseek-ai/deepseek-v4-pro-0813. You call it through the same OpenAI-compatible endpoint as every other model and pay in rupees.

What DeepSeek V4 Pro (0813) is

DeepSeek released the 0813 build of V4 Pro on 13 August 2026. It is a very large mixture-of-experts model with well over a trillion total parameters, released as open weights under the MIT licence. DeepSeek lists support for a context of up to one million tokens. unoblox lists a context length of 1,048,576 tokens and tool calling support, with text in and text out.

The 0813 tag is a specific dated build, so pin to this exact model id if you need reproducible behaviour across releases.

Where it fits

• Long-context analysis across large repositories, logs or document sets.
• Tool-using agents that call functions over many steps.
• Heavy reasoning and drafting where you want an open-weight lineage rather than a closed model.

V4 Pro is the larger and costlier sibling. For high-volume, lighter work look at DeepSeek V4 Flash (0731), which is priced far lower per token. A common pattern is to start on Flash and move specific hard cases to Pro.

₹ pricing for DeepSeek V4 Pro (0813)

On unoblox, DeepSeek V4 Pro (0813) costs ₹125.17 per million input tokens and ₹250.33 per million output tokens. That displayed price is what you are billed per token; nothing is added on top. Cached input tokens are listed at ₹9.63 per million. The model's context length on unoblox is 1,048,576 tokens.

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
DeepSeek V4 Pro (0813)	₹125.17	₹250.33
DeepSeek V4 Flash (0731)	₹5.78	₹17.33
DeepSeek V3.2	₹25.03	₹36.59
DeepSeek V4.1 Flash	₹20.22	₹60.66

As a worked example, a month with 20 million input tokens and 4 million output tokens comes to about ₹3,504.72 on DeepSeek V4 Pro (0813) (₹2,503.40 for input plus ₹1,001.32 for output). Live rates are always on the model page, and each request is rounded up to the nearest paisa.

Calling DeepSeek V4 Pro (0813) from the unoblox endpoint

unoblox is OpenAI-compatible, so any OpenAI SDK works by changing the base URL and key.

curl https://api.unoblox.ai/v1/chat/completions \
  -H "Authorization: Bearer YOUR_UNOBLOX_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "deepseek-ai/deepseek-v4-pro-0813",
    "messages": [{"role": "user", "content": "Summarise this support ticket in two lines."}],
    "max_tokens": 512
  }'

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_UNOBLOX_API_KEY",
    base_url="https://api.unoblox.ai/v1",
)
resp = client.chat.completions.create(
    model="deepseek-ai/deepseek-v4-pro-0813",
    messages=[{"role": "user", "content": "Summarise this support ticket in two lines."}],
    max_tokens=512,
)
print(resp.choices[0].message.content)

Get a key at https://unoblox.ai/sign-in and top up your wallet in rupees.

Notes on parameters, billing and data

Use max_tokens to cap output and the usual sampling parameters listed for the model. The model is served by an external provider rather than from India, so do not assume data residency. unoblox contributes rupee pricing, a GST invoice and a unified endpoint.

Frequently asked questions

Is DeepSeek V4 Pro (0813) available in India? Yes, as deepseek-ai/deepseek-v4-pro-0813 on https://api.unoblox.ai/v1, billed in rupees.

What is the context length? unoblox lists 1,048,576 tokens.

Is the model open weight? DeepSeek released the V4 family with open weights under the MIT licence.

Does it support tool calling? Yes, the unoblox catalogue lists tools and tool_choice among its supported parameters.

How does it differ from V4 Flash? Pro is the larger flagship; Flash is the faster, lower-priced tier. Both are on unoblox.

Is the data kept in India? No claim is made. Inference runs at the model provider; only billing and invoicing are India-based.

Get started in rupees → https://unoblox.ai/sign-in