DeepSeek V4 Flash (0731) API in India

DeepSeek V4 Flash (0731) is the dated July 2026 build of DeepSeek's fast, economical V4 tier. It is live on unoblox as deepseek-ai/deepseek-v4-flash-0731, with rupee billing and a GST invoice.

What DeepSeek V4 Flash (0731) is

DeepSeek released the 0731 build on 31 July 2026 as open weights under the MIT licence. It is a large mixture-of-experts model of roughly 300 billion total parameters, with the same architecture and size as the earlier Flash preview but post-trained again. unoblox lists a context length of 1,048,576 tokens, reasoning support and tool calling, with text in and text out.

It keeps the Flash identity: lots of context and strong cost efficiency, rather than the biggest-model ceiling.

Where it fits

• Chat and RAG backends that run thousands of requests an hour.
• Agent loops where cost per step adds up quickly.
• Long-context summarisation where a million-token window avoids chunking.

This is a sensible default for volume. If you already use the earlier deepseek-ai/deepseek-v4-flash, test the 0731 build on your prompts before switching, since it is a distinct model id with its own price. Step up to V4 Pro (0813) for the hardest cases.

₹ pricing for DeepSeek V4 Flash (0731)

On unoblox, DeepSeek V4 Flash (0731) costs ₹5.78 per million input tokens and ₹17.33 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 ₹1.44 per million. The model's context length on unoblox is 1,048,576 tokens.

Model	Input (₹ / 1M tokens)	Output (₹ / 1M tokens)
DeepSeek V4 Flash (0731)	₹5.78	₹17.33
DeepSeek V4 Flash	₹9.10	₹18.20
DeepSeek V4.1 Flash	₹20.22	₹60.66
DeepSeek V4 Pro (0813)	₹125.17	₹250.33

As a worked example, a month with 20 million input tokens and 4 million output tokens comes to about ₹184.92 on DeepSeek V4 Flash (0731) (₹115.60 for input plus ₹69.32 for output). Live rates are always on the model page, and each request is rounded up to the nearest paisa.

Calling DeepSeek V4 Flash (0731) 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-flash-0731",
    "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-flash-0731",
    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

Cap output with max_tokens. The model advertises reasoning support in the catalogue. It runs at an external provider and not in India, so we do not claim data residency. unoblox provides rupee pricing, GST invoicing and one endpoint.

Frequently asked questions

What does the 0731 mean? It marks the build date, 31 July 2026. Use the full model id to pin it.

Is it available in India? Yes, deepseek-ai/deepseek-v4-flash-0731 is on https://api.unoblox.ai/v1.

How long a context can I use? unoblox lists 1,048,576 tokens.

Does it support tools and reasoning? The catalogue lists both tools and reasoning for this model.

How does it differ from the older V4 Flash id? They are separate model ids with separate prices. Compare them on your workload.

Do I get a GST invoice? Yes, usage is billed in rupees with GST.

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