See what each AI model will actually cost you per month in rupees. Enter your tokens and request volume — the cheapest model for your workload floats to the top.
~15.0M tokens/month. Tip: 1 token ≈ 0.75 words.
Estimates use approximate published API rates (USD per 1M tokens, June 2026) converted at your ₹/$ rate — always confirm live pricing with the provider. Open-weight models (Llama 4, Qwen, Gemma, and these DeepSeek/Mistral models) are also free to self-host, paying only for hardware.
Want a full quality-vs-price view?
Compare these models on benchmarks, context window and weights side by side.
Compare models