Sovereign AI
Sovereign programme record · India · 2024–2029
India: ~$1.25B announced · disbursement not on record
India: IndiaAI Mission
- Announced~$1.25B
- Government funding~$1.25B
- Committed~$1.25B
- Disbursed—
Funding committed. 100% of government-only announced funding committed; no disbursement recorded. Announced in INR, converted at 82.78 as of 2024-03-07 via Board of Governors of the Federal Reserve System, H.10 via FRED DEXINUS.
Hardware on record
| Model | Count | Status | Organisation / role |
|---|---|---|---|
| GPUs (vendor breakdown undisclosed) | >38,000 | Onboarded for shared access (PIB, October 2025) | Compute providers accessed through IndiaAI; ownership not specified |
Shared access: Accelerators onboarded for subsidised shared access as of October 2025; vendor composition is unspecified. Primary source · 2025-10-12 | |||
| Google Trillium TPU | — | Third-round tender (August 2025); deployment not established | Compute providers accessed through IndiaAI; ownership not specified |
Empanelled partners · 14
| Partner | Accelerators | Model | Stage | Source |
|---|---|---|---|---|
| Yotta Data Services | 15,080 | NVIDIA H100 + NVIDIA B200 Blackwell | Empanelled | Research database · 2025-06-24 |
Shared access: Yotta’s June 2025 allocations: 6,888 H100 plus 8,192 B200; delivery was incomplete. Reporting · 2025-06-24 The June 2025 report counts 6,888 H100s allocated to Yotta, with 2,592 still due for delivery, plus 8,192 B200s selected for second-round deployment. | ||||
| Jio Platforms | 1,020 | NVIDIA H200 + AMD MI300X | Empanelled | Research database · 2025-06-24 |
Shared access: Jio’s June 2025 procurement: 752 H200 plus 268 MI300X accelerators. Reporting · 2025-06-24 Jio was procuring 752 H200s and 268 AMD MI300X accelerators in June 2025, according to Indian Infrastructure. | ||||
| Tata Communications | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names Tata among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| CMS Computers | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names CMS among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| CtrlS Datacenters | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names CtrlS among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| E2E Networks | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names E2E among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| Locuz Enterprise Solutions | 1,300 | NVIDIA H100 | Empanelled | Research database · 2025-08-18 |
Shared access: Locuz’s additional third-tender H100 offer in August 2025. Reporting · 2025-08-18 Inc42 reports an additional 1,300 H100s from Locuz in the August 2025 third tender. This row covers that round; the report does not establish installation. | ||||
| Orient Technologies | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names Orient among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| NxtGen Datacenter and Cloud Technologies | — | — | — | Primary source · 2024-12-03 |
PIB’s December 2024 notice names NxtGen among the bidders awaiting technical evaluation; it gives no allocation or installation count. | ||||
| Vensysco Technologies | 2,300 | NVIDIA H100 + AWS Trainium 1 + AWS Inferentia 2 | Empanelled | Research database · 2025-08-18 |
Shared access: Vensysco’s second-round bid: 2,000 H100, 100 Trainium1 and 200 Inferentia2 accelerators. Reporting · 2025-05-15 Vensysco’s second-round bid covered 2,000 H100s, 100 Trainium1 units, and 200 Inferentia2 units. The 2,300 total includes non-GPU accelerators; the later third-round report concerns pricing revisions. | ||||
| Netmagic (NTT Global Data Centers) | — | — | Shortlisted | Research database · 2025-04-22 |
Inc42 reports Netmagic’s technical qualification in the second round. The cited report gives no allocation count. | ||||
| Cyfuture India | 1,184 | NVIDIA H100 + L40S + A100, AMD MI300 + MI325, Intel Gaudi 2 + Gaudi 3 | Shortlisted | Research database · 2025-05-15 |
Shared access: Cyfuture’s mixed-accelerator bid in the May 2025 second tender. Reporting · 2025-05-15 Elets reports a second-round bid for 1,184 accelerators in May 2025. A bid does not establish the number installed or available to users. | ||||
| Sify Digital Services | 2,500 | Google Trillium TPU + NVIDIA H200 + NVIDIA L4 | Empanelled | Research database · 2025-08-18 |
Shared access: Sify’s third-tender offer: 1,000 Trillium TPUs, 800 H200 and 700 L4 GPUs. Reporting · 2025-08-18 Sify’s August 2025 third-tender offer comprised 1,000 Trillium TPUs, 800 H200s, and 700 L4s. The 2,500 units combine TPUs and GPUs; installation was not established by the report. | ||||
| Ishan Infotech | 50 | Google Trillium TPU | Empanelled | Research database · 2025-08-18 |
Shared access: Ishan’s third-tender offer of 50 Trillium TPUs in August 2025. Reporting · 2025-08-18 Ishan’s August 2025 third-tender offer included 50 Trillium TPUs, part of the 1,050 TPUs reported for that round. The report does not establish installation. | ||||
External dependence
NVIDIA hardware: High · US technology: High. These are editorial assessments based on the programme’s procurement disclosures.
Named partners: MeitY, Digital India Corporation, NVIDIA, Google, G42, L&T, Yotta.
IndiaAI pays for access to computing through empanelled service providers and supports datasets, models, training, and startup finance. The Cabinet approved ₹10,371.92 crore over five years. In October 2025, PIB reported more than 38,000 GPUs onboarded for shared access; it gave no government-owned fleet count or vendor breakdown.
Data-quality caveats on this record · 10
- The Cabinet’s ₹10,371.92 crore approval covers the whole IndiaAI Mission. It is not a GPU-purchase budget.
- The compute programme uses a public-private partnership. The government’s aggregate GPU count does not establish how many devices it owns or how much capacity each user can reserve.
- Partner records span tender rounds, bids, and installation reports. They also include Google TPUs and AWS Trainium and Inferentia accelerators, so summing their counts does not reconstruct the government’s GPU total.
- The August 2025 third-tender report includes 1,050 Trillium TPUs within 3,850 additions to the existing 34,333-unit pool. Those TPUs are not added to the government’s later rounded 38,000 count; the sources do not establish a separate GPU-only total.
- IndiaAI received ₹21.79 crore in 2024–25 and ₹379.15 crore in 2025–26 through 9 February 2026, according to the parliamentary answer of 13 February.
- A separate parliamentary report records actual utilisation of ₹19.24 crore in 2024–25 and ₹256.86 crore in 2025–26 through 31 December 2025. These are expenditure figures at an earlier date than the reported fund transfers.
- The dollar estimate converts the Cabinet’s rupee allocation using 82.78 INR per USD on 7 March 2024, from the Federal Reserve’s H.10 series distributed by FRED. The displayed $1.25B is rounded.
- G42 and Cerebras announced an 8-exaFLOP system for India in February 2026. The announcement describes a future deployment and does not specify the precision needed to compare that performance with other systems.
- NVIDIA describes private expansion plans including L&T’s 30 MW Chennai and 40 MW Mumbai facilities and Yotta’s Blackwell Ultra cloud. These plans are separate from the mission’s government allocation.
- Adani and Jabil announced their intention to form an AI-hardware manufacturing alliance in June 2026, with formal documentation still in preparation. The release gives no alliance-specific investment amount or operating compute capacity.
The evidence trail
20 cited sources- Digital India’s July 2026 account of GPUs onboarded for shared compute access. — Primary source · accessed 2026-07-14
- Adani’s June 2026 announcement of an intended hardware-manufacturing alliance with Jabil, with formal documentation still in preparation. — Primary source · accessed 2026-07-14
- IndiaAI’s overview of the compute-access programme. — Primary source · no accessed date on record
- DD News account of the mission allocation and aggregate GPU capacity. — Primary source · no accessed date on record
- NVIDIA descriptions of private cloud infrastructure and model-development projects in India. — Primary source · no accessed date on record
- MediaNama’s April 2026 account of the February parliamentary answer on funds received by IndiaAI. — Research database · accessed 2026-04-08
- Federal Reserve H.10 INR/USD exchange-rate series, distributed by FRED. — Primary source · accessed 2026-05-08
- PIB’s March 2024 statement that Cabinet approved more than ₹10,300 crore for IndiaAI. — Primary source · accessed 2026-05-08
- Indian Infrastructure’s June 2025 account of first- and second-round provider installations. — Research database · accessed 2026-05-08
- Inc42’s August 2025 report on the third compute tender, including GPU and TPU counts. — Research database · accessed 2026-05-08
- Elets report on second-round bids from compute providers, May 2025. — Research database · accessed 2026-05-08
- PIB’s 3 December 2024 notice listing 19 bidders before technical evaluation. — Primary source · accessed 2026-05-08
- PIB’s Cabinet decision approving ₹10,371.92 crore for IndiaAI on 7 March 2024. — Primary source · accessed 2026-03-28
- IndiaAI’s account of Cabinet approval for the mission. — Primary source · accessed 2026-03-28
- Business Standard reporting on the proposed share of mission funds for compute procurement. — Press or analyst · accessed 2026-03-28
- PIB’s October 2025 account of more than 38,000 GPUs onboarded for subsidised access. — Primary source · accessed 2026-09-08
- G42’s February 2026 announcement of a planned system for India, described as 8 exaFLOPS without a stated precision. — Primary source · accessed 2026-09-08
- The Economic Times’ August 2025 interview with IndiaAI’s CEO about the third tender, including 1,050 TPUs within the 3,850-unit addition. — Press or analyst · accessed 2026-09-08
- Rajya Sabha answer of 13 February 2026, page 2: annual funds received by IndiaAI through 9 February, with the programme budget and cloud-access arrangements. — Primary source · accessed 2026-09-08
- Parliamentary committee’s March 2026 report on MeitY’s 2026–27 demands for grants, paragraph 51: IndiaAI actual utilisation through 31 December 2025. — Primary source · accessed 2026-09-08
Machine accessThe same record, machine-readable: the evidence-trail feed and the derivation-chain export.