DeepSeek Pricing in India (₹)

DeepSeek model pricing in rupees

DeepSeek offers cost-effective open-weight reasoning models. unoblox brings them to India at transparent per-token pricing, billed entirely in rupees on a GST invoice.

DeepSeek pricing (per 1M tokens)

Model	Input ₹	Output ₹	Context	Best for
DeepSeek V4 Flash	₹9.07	₹18.14	64K	Fast, cheap inference
DeepSeek V4.1 Flash	₹20.16	₹60.48	128K	Extended reasoning
DeepSeek V3.2	₹26.21	₹38.30	64K	Strong coding, math

All rates are indicative. Live rates on model pages at /models.

How pricing works

• No markups. unoblox passes through provider rates, adding only platform fees (typically 5–10%).
• Per-token billing. You're charged for input + output tokens in your request. No minimum spends, no overage fees.
• Monthly GST invoice. All charges billed in rupees on a single invoice with full tax credit eligibility.
• Transparent settlement. See usage & costs in real-time on your dashboard.

Cost comparison

For a typical task (1000 input tokens, 500 output tokens):

Model	Input cost	Output cost	Total
DeepSeek V4 Flash	₹0.009	₹0.009	₹0.018
DeepSeek V4.1 Flash	₹0.020	₹0.030	₹0.050
DeepSeek V3.2	₹0.026	₹0.019	₹0.045

At the rates above. Actual costs vary based on live model pricing.

When to use each model

DeepSeek V4 Flash: Fastest, cheapest. Use for chatbots, summarization, simple classification. Ideal for high-volume, latency-sensitive workloads.

DeepSeek V4.1 Flash: Extended reasoning with 128K context. Use for document Q&A, long-form content analysis, multi-step reasoning. Cost is slightly higher but context is double.

DeepSeek V3.2: Strong coding and mathematical reasoning. Use for code generation, data transformation, complex problem-solving.

Pricing FAQ

Q: How do I start using DeepSeek on unoblox? Sign up, create an API key, and point your OpenAI SDK at https://api.unoblox.ai/v1 with model deepseek-ai/deepseek-v4-flash. See quickstarts at /models/deepseek-ai/deepseek-v4-flash.

Q: Do you offer volume discounts? Custom rates are available for enterprise-scale usage — contact sales to discuss your monthly volume.

Q: What's the difference between V4 and V4.1 Flash? V4 Flash is faster; V4.1 Flash has double the context (128K). Choose based on latency vs. context trade-off.

Q: Can I use DeepSeek for production? Yes. Full SLA coverage (99.95% uptime), logs retained 90–180 days. See production compliance at /guides/ai-api-uptime-sla-india.

Q: Is my data secure and resident in India? Yes. Data residency guaranteed in India. No third-party access. See our security & compliance page.

Q: Can I query live rates programmatically? Yes, via /v1/models. Returns pricing for all available models in your account.

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