Mistral API in India

Mistral AI's models are built for efficiency, and they're now reachable in India through unoblox's OpenAI-compatible endpoint — ₹-native pricing, a monthly GST invoice, and no international payment friction to sort out first.

Why Mistral matters

Mistral has a reputation for strong token-per-rupee value. Its models are:

• Lean and fast — comparable accuracy to larger models at lower latency.
• Open and documented — Mistral publishes detailed model cards and benchmark results.
• Multilingual — strong across European languages and English.
• Commercial-friendly — clear licensing terms for enterprise use.

Pricing and availability

Current Mistral pricing in rupees is live on /models — new Mistral releases are added regularly, so that's the page to check before committing a workload rather than a number fixed here.

Choosing a Mistral tier

Priority	Look for	Why
Lowest cost, simple tasks	The smallest listed tier	Fastest, cheapest per token
Balanced reasoning and speed	A mid-size tier	Good default for most production chat
Highest quality	The largest listed tier	Closest to frontier-model reasoning

Exact model names and live ₹ rates for each tier are on /models.

Integration in 90 seconds

from openai import OpenAI

client = OpenAI(api_key="ub-gw-...", base_url="https://api.unoblox.ai/v1")
response = client.chat.completions.create(
    model="mistral/[model-name]",  # exact id on /models
    messages=[{"role": "user", "content": "Hello, Mistral!"}],
)

Streaming, tool use, and JSON output are all supported.

Best for

• European teams entering India — Mistral's European roots plus rupee billing make compliance straightforward.
• Real-time chat — a latency profile well suited to user-facing apps.
• Cost-conscious teams — consistently strong value per rupee.
• Multilingual content — solid European-language support alongside English.

Mistral plus unoblox

One key, every model — route to Mistral, GPT, Claude, Qwen, or Llama from the same key, and test Mistral against competitors without separate accounts.

GST compliance — one monthly rupee invoice, input-tax credit claimable, nothing to reconcile against a foreign statement.

Always-current catalog — new Mistral models are added as they release; /models always reflects what's live.

Frequently asked questions

Where can I see exact Mistral pricing? /models — search "Mistral" to filter, and check the model ID you need before you integrate.

Does Mistral support vision? It depends on the specific model — check the card on /models, since some Mistral releases are multimodal and others are text-only.

Is Mistral good for production? Yes — plenty of European and Indian teams run Mistral in production, backed by unoblox's 99.95% platform SLA.

Can I fine-tune Mistral models? Custom fine-tuning isn't exposed through the API; Mistral offers its own on-premise licensing for teams that need that.

What if I need a Mistral model that's not listed? Email support@unoblox.ai with the model name — new providers and models are onboarded regularly, and /models reflects the current catalog.

How does Mistral's pricing compare to Llama or Gemma? Check /models for a live side-by-side — all three are open-weight-friendly options worth comparing on your specific workload.

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