One subscription.
Every top model.

Every plan starts with 48 hours free. Use the product fully — chat, attach PDFs, route to frontier models. If you decide it's not for you, message us and we refund your payment, less only the GST and payment-gateway fees that are set by third parties and not controlled by MultiRiz.

During the launch window (10 days, see the countdown banner), buying the 12-month plan charges you for 8 months — the other 4 are on the house. Available on Pro and Ultra annual billing.

Indian tax rules require 18% GST on SaaS subscriptions. Our headline prices are exclusive of tax, so 18% GST is added at checkout and itemised as a separate line on your invoice — this keeps the tax transparent and lets businesses claim input-tax credit.

Every prompt is scored 0–10 and routed by difficulty across four bands — Simple, Medium, Complex and Very-Complex — all through a single gateway, OpenRouter. Simple prompts auto-route to OpenRouter Free on every plan. Basic's paid bands use fast, ultra-low-cost models like Gemini 2.5 Flash Lite, GPT-4.1 Nano, DeepSeek V4 Flash, Qwen 3.5 Flash, DeepSeek V3 and MiniMax M2. Pro steps up to smarter mid-tier models with bigger context, including GLM 5.2, Kimi K3, GPT-4.1 Mini and Claude Haiku 4.5 for the hardest Pro prompts. Ultra unlocks our top model — Claude Sonnet 5 (200K context) — plus Gemini, GPT-5.1, Grok 4.3, DeepSeek R1 and more. Ultra also lets you pick any model yourself from the in-chat picker, or leave it on Auto smart-routing. Claude Sonnet 5 is the ceiling — we never route anything above it.

Web grounding is available on the Ultra plan. Ultra routes every answer through OpenRouter with live web search enabled — MultiRiz searches high-authority live sources and adds numbered, clickable citations under the reply. When sources disagree or fall short it says so instead of inventing details, which sharply cuts hallucinations. You can toggle web grounding off for any single message directly in the chat. Basic and Pro route through the same OpenRouter gateway but use the models' own training knowledge to answer accurately.

On annual (12-month) billing you can split your payment into easy monthly instalments at no extra cost. If your bank charges interest on the EMI, we absorb that interest entirely — you pay only the real plan price, never more. It's offered at checkout through Cashfree on Pro and Ultra annual plans, subject to your bank's or card's EMI eligibility.

AI Credits are a simple daily allowance that reflects how much computing your prompts use — they're not money and you can't buy them separately. Simple questions are always free (0 credits). For everything else, the credits a reply costs scale with how heavy the model and answer actually are, so a quick medium reply costs only a credit or two while a long, very-complex answer on a top model costs more. You only ever see your daily credits and messages, both reset at midnight with no rollover, and credits are counted only after a successful reply — if something fails, you're charged nothing. On Ultra, the in-chat model picker shows each model's credit cost before you choose.

On Basic and Pro, MultiRiz automatically routes each prompt to the best-fit model. Ultra adds full control: a model picker in the chat lets you select any model manually, or keep Auto smart-routing. You get 60 premium messages/day and 200 AI Credits/day — our most generous allowance — with fair-use limits so premium models stay sustainable. Bill it monthly or use the Pro & Ultra annual offer to get 4 months free.

Yes — one click from your dashboard. No lock-in, no cancellation fees. If you cancel inside the trial window the charge is fully refundable after deducting the GST and payment gateway fee, as these are not controlled by MultiRiz.

UPI, credit/debit cards and net banking through Cashfree — every popular Indian payment method.

Email support@multiriz.com with your team size and expected volume. We typically reply within one business day with a tailored quote and SSO/SAML details.

Yes. Conversations are encrypted in transit (TLS) and at rest, never sold, and never used to train models. When you attach a PDF or text file, the text is extracted in your browser and only that text is sent to the model; images and other files are sent over an encrypted connection to our servers and passed to your selected model to ground its answer. We keep uploaded content while your account is active and delete it when you delete the chat or your account, subject to short-term backups. See our Privacy Policy for full details.