# Tensor Machine — Full Text > India's sovereign AI inference platform — an OpenAI-compatible API served from Indian data centres, priced per-token in INR, DPDP-compliant, content logging off by default. Index: https://tensormachine.ai/llms.txt · Docs full text: https://docs.tensormachine.ai/llms-full.txt # Models > The frontier and open-weight models available through Tensor Machine — one OpenAI-compatible API, priced per 1M tokens in INR (GST added at billing). 10 models listed. | Model ID | Maker | Context | Type | Input ₹/1M | Output ₹/1M | | --- | --- | --- | --- | ---: | ---: | | tm/glm-5-2 | Z-AI | 1M | Chat, Code | ₹157.65 | ₹472.96 | | tm/gpt-oss-120b | OpenAI | 131K | Chat | ₹4.20 | ₹21.02 | | tm/qwen3-5-122b | Alibaba | 256K | Chat, Code, Vision | ₹52.55 | ₹367.86 | | tm/qwen3-5-9b | Alibaba | 256K | Chat | ₹15.77 | ₹21.02 | | tm/minimax-m3 | MiniMax | 1M | Chat, Code, Vision | ₹42.04 | ₹210.21 | | tm/deepseek-v4-pro | DeepSeek | 1M | Chat, Code | ₹183.93 | ₹367.86 | | tm/kimi-k2-6 | MoonshotAI | 256K | Chat | ₹105.10 | ₹420.41 | | tm/kimi-k2-7-code | MoonshotAI | 256K | Code | ₹131.38 | ₹472.96 | | tm/deepseek-v4-flash | DeepSeek | 1M | Chat | ₹15.77 | ₹31.53 | | tm/qwen3-235b | Alibaba | 128K | Chat, Code | ₹7.36 | ₹48.35 | Full model library: https://tensormachine.ai/models · Pricing & cost calculator: https://tensormachine.ai/pricing --- # Pricing > Usage-based, per-token AI inference pricing in INR — GST-compliant invoices, no minimums. Top up a prepaid wallet and pay only for the tokens you send. ## Self-serve (per-token) Per-token pricing, no foreign exchange, GST added at billing (18%). Cached input is billed at the input rate unless a discounted cached rate is shown. | Model | Maker | Context | Input ₹/1M | Output ₹/1M | | --- | --- | --- | ---: | ---: | | GLM-5.2 | Z-AI | 1M | ₹157.65 | ₹472.96 | | GPT-OSS 120B | OpenAI | 131K | ₹4.20 | ₹21.02 | | Qwen3.5 122B | Alibaba | 256K | ₹52.55 | ₹367.86 | | Qwen3.5 9B | Alibaba | 256K | ₹15.77 | ₹21.02 | | MiniMax-M3 | MiniMax | 1M | ₹42.04 | ₹210.21 | | DeepSeek-V4 Pro | DeepSeek | 1M | ₹183.93 | ₹367.86 | | Kimi K2.6 | MoonshotAI | 256K | ₹105.10 | ₹420.41 | | Kimi K2.7 Code | MoonshotAI | 256K | ₹131.38 | ₹472.96 | | DeepSeek-V4 Flash | DeepSeek | 1M | ₹15.77 | ₹31.53 | | Qwen3-235B | Alibaba | 128K | ₹7.36 | ₹48.35 | ## Enterprise Two ways to work with us beyond self-serve. Tell us your volume and requirements and we'll put a quote together. Committed-volume rate cards are real; we don't advertise an automatic percentage-tier grid. ### Scale For teams whose volume is growing fast and who want the rate to reflect it. - Committed-volume & sustained-use discounts - Prepaid wallet or monthly invoicing - Higher rate limits and per-team quotas - Usage analytics across teams - GST-compliant INR invoices ### Enterprise For organisations that need the controls, contracts, and support to standardise on us. - SSO (SAML / OIDC) and SCIM provisioning - Data Processing Agreement (DPA) - Audit logs with SIEM export - NET payment terms on invoices - Priority support Full pricing page with a cost calculator: https://tensormachine.ai/pricing · Talk to us: https://tensormachine.ai/contact --- # Tensor Machine Is Live: Sovereign AI Inference, Built for Bharat > After two years of building in stealth, India's first dedicated sovereign AI inference platform is live today. No prompt data leaves Indian borders. No foreign exchange markup. No DPDP grey areas. Here is what we built, and why it matters for Indian enterprises right now. - Author: Tensor Machine - Published: 2026-01-15 - Tag: Announcement - URL: https://tensormachine.ai/blog/tensor-machine-live-sovereign-ai-inference --- ## The Problem We Set Out to Solve For years, Indian engineering teams have had no good option for AI inference. The only serious providers were American. Your data travelled to servers in Virginia or Oregon every time a user hit your AI feature. Your finance team dealt with USD invoices subject to 18% GST reverse charge. Your legal team worried about DPDP Act implications and data residency requirements that foreign providers could not credibly address. ## What Tensor Machine Delivers Tensor Machine is India's first sovereign AI inference platform, purpose-built for organisations that cannot afford data residency compromises. Every request is processed inside Indian data centres. No data is retained after inference. No training happens on your data. DPDP Act 2023 compliant by architecture, not by contract. We offer full OpenAI-compatible API access, so migration is a one-line change to your existing codebase. INR billing with GST-compliant invoices. Data Processing Agreements available for enterprise customers. ## Frontier Models, Available Now From DeepSeek V3 and Kimi K1.5 to Minimax and GLM, we have curated a catalogue of the world's most capable open-weight models. All running on India's sovereign infrastructure. All priced in INR. All accessible through a single API key. ## Who Should Use Tensor Machine We built Tensor Machine for three types of customers. Regulated enterprises in BFSI, healthcare, and government who need DPDP compliance and data residency guarantees. Product teams building AI features who want OpenAI API compatibility without USD billing complexity. AI-native startups who want the best models at the lowest price, billed in INR with proper GST invoices. ## Get Started Today Create a free account at tensormachine.ai. Your first API key is ready in under two minutes. No credit card required for the free tier. If you are evaluating for enterprise use, reach out to our team for a compliance briefing and DPA. --- # ₹80 Lakhs vs ₹18 Lakhs: The True Cost of AI Inference for Indian Teams > We modelled the total cost of AI inference for a mid-sized Indian SaaS company processing one billion tokens per month. US-based providers cost ₹80 lakhs all-in once you account for FX exposure, GST reverse charge, and compliance overhead. The same workload on Tensor Machine costs ₹18 lakhs. Here is the full breakdown. - Author: Tensor Machine - Published: 2025-12-10 - Tag: Analysis - URL: https://tensormachine.ai/blog/ai-inference-cost-comparison-india-inr --- ## The Numbers Nobody Talks About When Indian engineering leaders evaluate AI inference costs, they typically compare headline USD prices. They convert to INR and move on. That analysis misses three significant cost categories that are unique to Indian buyers of foreign SaaS — and together they can double the effective cost of running AI at scale. ## Hidden Cost 1: Foreign Exchange Risk The INR has depreciated against the USD by an average of 3–4% per year over the last decade. An AI workload budgeted in INR at the start of the year costs more in rupee terms by year-end — purely due to currency movement. For fast-growing AI workloads, the FX exposure compounds with volume. Teams that forecast AI spend in INR but pay invoices in USD face a structural budget risk with no easy hedge. ## Hidden Cost 2: GST Reverse Charge Under India's GST framework, software services imported from foreign providers are subject to 18% GST under the reverse charge mechanism. Even for companies that can claim full input tax credit, the cash flow impact of paying GST upfront on every invoice adds meaningful administrative burden — and for companies under the composition scheme, it is a direct cash cost. ## Hidden Cost 3: Compliance Overhead Regulated industries — BFSI, healthcare, government — must conduct vendor assessments for every foreign AI provider handling personal data. Data Protection Agreements, security audits, legal reviews, and ongoing vendor monitoring. For a typical regulated enterprise, this costs ₹15–25 lakhs in the first year and ₹5–8 lakhs annually thereafter. ## The Model: 1 Billion Tokens Per Month For a typical Indian B2B SaaS product with 50,000 active users, one billion tokens per month is a realistic inference volume. This covers AI search, document summarisation, customer support automation, and coding assistance features. At current frontier model pricing of approximately ₹420 per million blended tokens (equivalent to $5/M at ₹84 per dollar), processing one billion tokens costs ₹4.2 lakhs per month in base compute. Add 18% GST reverse charge (₹75,600), FX exposure at 3.5% annualised (₹12,250 per month), and compliance overhead amortised at ₹1.5 lakhs per month for a regulated enterprise — and the all-in cost reaches approximately ₹6.6 lakhs per month, or ₹79–80 lakhs per year. Tensor Machine's equivalent offering at ₹150 per million blended tokens costs ₹1.5 lakhs per month for the same one billion tokens. GST included. INR invoiced. Zero compliance overhead for DPDP-regulated use cases. Annual cost: ₹18 lakhs. Saving of over ₹60 lakhs per year. ## Why INR Billing Changes Everything The savings are significant. But the more important shift is structural. When you pay in INR with GST-compliant invoices, your AI spend becomes a normal line item in your P&L — forecastable, auditable, and manageable. No FX exposure. No reverse charge complexity. No foreign vendor assessments every quarter. Your AI budget behaves like your cloud budget. --- # DPDP Act 2023 and Your AI Product: What Engineering Teams Actually Need to Know > India's Digital Personal Data Protection Act 2023 has direct implications for any product that processes personal data through AI inference. Most engineering teams do not know where their prompts go, who processes them, and whether they qualify as personal data under Indian law. This guide breaks it down in plain terms. - Author: Tensor Machine - Published: 2025-12-05 - Tag: Compliance - URL: https://tensormachine.ai/blog/dpdp-act-2023-ai-inference-developers-guide --- ## What the DPDP Act Actually Is The Digital Personal Data Protection Act 2023 received Presidential assent on 11 August 2023. It is India's first comprehensive data protection legislation and creates a framework for how personal data of Indian residents must be collected, stored, processed, and transferred. The Act applies to any organisation that processes digital personal data in India or processes data of Indian residents from outside India — including SaaS companies, AI platforms, and their downstream customers. ## How AI Inference Touches Personal Data Here is where engineering teams often get caught out. A prompt sent to an AI model can contain personal data. A customer support prompt may include a user's name, account number, or medical history. A document summarisation prompt may include employee records or client contracts. Under Section 8 of the DPDP Act, if you transfer this data to a foreign AI provider, you must ensure your Data Processor meets the Act's obligations — and under Section 16, cross-border transfers of personal data are subject to the restrictions notified by the Central Government. The DPDP Act defines personal data as any data about an identifiable natural person. Prompts containing names, contact details, financial information, health data, or combinations that could identify an individual qualify. Whether embeddings derived from personal data qualify depends on whether the embedding can reasonably be used to identify the individual — a point legal counsel should assess for your specific use case. ## Data Fiduciary vs Data Processor — Where AI Providers Fit The DPDP Act distinguishes between Data Fiduciaries — entities that determine the purpose and means of processing personal data — and Data Processors, who process data on behalf of Fiduciaries. If you build an AI product, you are likely a Data Fiduciary. Your AI inference provider is a Data Processor. Under Section 8(2), Data Fiduciaries are responsible for ensuring their Data Processors comply with the Act's provisions, including through contractual obligations. ## What Compliance Actually Requires For most AI-enabled products, DPDP compliance requires four things. First: a lawful basis for processing — typically consent obtained in the manner prescribed under Section 6, or a legitimate use recognised under Section 7. Second: a Data Processing Agreement with your AI inference provider committing them to process data only as directed, maintain appropriate security safeguards, and not use data for any other purpose including model training. Third: implementing data localisation to the extent required by notifications under Section 16. Fourth: the ability to honour Data Principal rights — right to access, correction, erasure, and grievance redressal — which requires traceability of where personal data was processed. For regulated industries specifically — BFSI regulated by RBI, insurance by IRDAI, healthcare — sector-specific guidelines effectively require AI inference to happen within India for data classified as sensitive. The RBI circular on outsourcing of IT services (2023) and IRDAI guidelines on cloud adoption both require data to remain within Indian jurisdiction for regulated workloads. This is an audit point, not just best practice. ## A Practical Architecture Checklist For your engineering and product teams: confirm your AI inference provider processes all data within India. Obtain a signed Data Processing Agreement before your next vendor audit. Establish what your provider retains by default, and whether retention is opt-in or opt-out — ask about logs, embeddings, and conversation history separately, and get the default in writing. Map which product features send personal data to AI and which do not. Implement consent flows for AI features that process identifiable user data. Document your AI vendor assessments and keep them current for your Data Protection Officer. ## How Tensor Machine Helps Tensor Machine is designed from the ground up for DPDP compliance. All inference runs in Indian data centres. Content logging is off unless your organisation turns it on, and the router enforces that at the edge — retention is opt-in, not something you must remember to disable. We do not train on customer data and we do not sell it; the model provider serving a request necessarily receives the prompt, and no one else does. Data Processing Agreements are available to all paid customers. Our enterprise SLA includes compliance documentation packages for DPOs and legal teams, including evidence of data localisation and security controls. If you are evaluating AI inference providers for a regulated use case — BFSI, healthcare, or government — we are happy to walk through our compliance posture in detail, including audit support. Reach out to our enterprise team.