As of 31 December 2025, India's cumulative installed BESS capacity stood at roughly 1,082 MWh (about 1.08 GWh), per Mercom India's year-end tracking. Of that small base, solar-plus-storage hybrids accounted for more than 54%, solar-plus-wind round-the-clock (RTC) configurations for roughly 23%, and standalone BESS for about 21%, per the same tracking as reported by pv-magazine India in March 2026. Worth saying plainly: this is Mercom's own commissioned-capacity series, internally consistent from one publisher across successive quarters, but this piece could not independently confirm it against a CEA or MNRE installed-capacity dashboard, because both cea.nic.in and mnre.gov.in returned blocked-egress errors during this piece's research. Treat the Mercom series as the best available approximation of what is actually built, sourced secondarily rather than read directly off a government system.

A lithium-ion battery pack — the hardware India's BESS tenders are buying at record pace, and which its ACC PLI scheme has so far commissioned only 2.8% of the domestic cell capacity to build. Source: RudolfSimon, Wikimedia Commons, CC BY-SA 3.0.

From that base, the market accelerated sharply. Q1 2026 alone added 4.6 GWh, taking cumulative installed capacity to roughly 5.9 GWh by end-March 2026 — a reported 941% jump over the 442.7 MWh added in Q4 2025. By 30 June 2026, cumulative installed capacity reached about 9.3 GWh, with 8.2 GWh added in the first half of 2026 alone, described by Mercom as an "84-fold" year-on-year increase against the 98.4 MWh added in H1 2025. Read those percentages for what they are: real, large multiples, but multiples of a base so small a year earlier that even a modest number of large tenders reaching commercial operation can move the year-on-year figure by tens of times.

Pumped storage hydro (PSP), the country's older grid-storage technology, tells a slower-moving story alongside this. Installed PSP capacity reached 7.2 GW by March 2026, of which 5.7 GW was operational, per the same Mercom reporting. On a tendered-energy basis, PSP still holds the larger historical share — roughly 132 GWh PSP against 92 GWh BESS cumulatively tendered through December 2025, per India ESA's (IESA) year-end tracker — reflecting PSP's much longer development history against BESS's emergence as a tenderable, biddable asset class only since 2022. But near-term installation momentum is now overwhelmingly BESS-led, simply because a BESS project can go from tender to commissioning in 12–24 months where PSP civil works run for years; NTPC, NHPC and SJVN are all pursuing multi-gigawatt PSP pipelines (more on that below), but those additions arrive on a much longer clock than the BESS numbers above.

Tendering activity tells the same acceleration story from a different angle: 2025 alone saw 69 storage tenders totalling 102 GWh floated, nearly matching the combined tendered volume of 2018 through 2024, per IESA-linked reporting carried by pv-magazine India in January 2026. Mercom separately reports roughly 28 GW of storage tenders floated in H1 2026, with about 9 GW auctioned. Government's own forward-looking targets — spread across at least three separate CEA and National Electricity Plan documents of different vintages — all point toward something like a tenfold expansion of required storage capacity by the early 2030s versus the mid-2020s. The exact number depends heavily on which document is cited, which is exactly the kind of distinction Section 5 below picks up rather than glossing over.

The Ministry of Power's National Framework for Promoting Energy Storage Systems, issued in August 2023, sets the umbrella objectives: enabling round-the-clock dispatchable renewable power, redesigning power markets to value storage services such as frequency regulation and voltage support, waiving Inter-State Transmission System (ISTS) charges for storage use, laying out BESS procurement rules, and folding energy storage systems into the infrastructure "harmonised master list" that affects financing eligibility and terms. Its existence and August 2023 date are corroborated across PIB, Renewable Watch and legal-compliance coverage, though this piece could not fetch the framework document itself, since powermin.gov.in and mnre.gov.in were not directly reachable in this research.

Underneath that framework sits an Energy Storage Obligation (ESO), notified by the Ministry of Power on 22 July 2022 alongside a Renewable Purchase Obligation trajectory: 1% of relevant procurement in FY2023-24, rising by roughly half a percentage point a year to 4% by FY2029-30, with the added condition that at least 85% of the energy stored annually in a system must come from renewable power for the obligation to count as fulfilled. A separate Ministry of Power order dated 10 June 2025 extended a 100% ISTS charge waiver to hydro PSP projects awarded on or before 30 June 2028 and to co-located BESS commissioned by the same date — for co-located BESS, conditional on the stored power being consumed outside the state where the project sits.

The two Viability Gap Funding (VGF) tranches are the piece of this machinery that has moved the most money. Tranche I was approved by the Union Cabinet on 6 September 2023; Tranche II, a materially different design, followed on 14 May 2026.

Both tranches per Cabinet decisions reported by PIB and secondary press (Energy-Storage.News, Energetica India, JMK Research); direct fetches of pib.gov.in were blocked in this piece's research, so figures above rest on multiple independently-worded secondary accounts of the same cabinet decisions, not a direct reading of the press release text.

Tranche II's state-wise allocation, per Energetica India and JMK Research's coverage, gave Rajasthan, Gujarat and Maharashtra the largest shares (4,000 MWh / 4 GWh each), with Karnataka and Andhra Pradesh at 1,500 MWh each. On the corporate pumped-storage side, NTPC's group is reported pursuing a 20 GW PSP pipeline aiming for 3–5 GW commissioned by FY2032, with its 1,000 MW Tehri PSP in Uttarakhand having begun commercial supply on its first unit; NHPC has a joint venture with AP Genco covering the 1 GW Yaganti and 800 MW Rajupalem projects and is reported planning 12.4 GW across Maharashtra and Andhra Pradesh; SJVN has signed a letter of intent for a 2.4 GW PSP at Darzo Nallah in Mizoram. On the tendering-structure side, IEEFA-JMK's 2025 analysis found standalone tenders made up more than 71% of total capacity tendered in India that year, with standalone BESS specifically accounting for 60% of that share — meaning standalone contracting has, for now, outpaced the RTC/FDRE (Firm and Dispatchable Renewable Energy) hybrid structures that combine generation and storage in one tender, at least by tendered volume in 2025. Those hybrid tenders price on a different metric (₹/kWh of delivered energy, not ₹/MW/month of capacity) — SECI's 1 GW FDRE-RTC tender discovered a lowest tariff of ₹5.25/kWh around August 2026, and REMCL's 1,000 MW RTC auction (awarded November 2025) went to ReNew Power, Bhalki Solar and Purvah Green among others — and that difference in metric is one more reason the tariff figures in the next section should not be read as describing the whole market.

Standalone BESS tariffs, quoted on a ₹/MW/month capacity-charge basis, have fallen almost in a straight line since India's first storage tender. SECI's first standalone BESS pilot, in August 2022, discovered a tariff of ₹10.83–10.84 lakh/MW/month. H2 2024 tenders came in around ₹3.0–3.2 lakh/MW/month. A January 2025 tender backed by VGF discovered ₹2.22 lakh/MW/month. An April 2025 two-hour system with no VGF subsidy at all came in at ₹2.8–2.85 lakh/MW/month, about 20% below the H2 2024 level. GUVNL's Gujarat auctions across 2025-26 cleared in the ₹2.10–2.30 lakh/MW/month range, with winners including H.G. Infra Engineering and, in a later phase, Advait Energy Transitions, Equentia Natural Resources and Sun Drops Energia. And in December 2025, SVR Electro Projects bid a national record low of ₹1.48 lakh/MW/month in APTRANSCO's 1,000 MW Andhra Pradesh tender — a two-hour system, the same configuration as the April 2025 tender above. One duration caveat worth keeping in view: four-hour systems have discovered tariffs around ₹2.85 lakh/MW/month even in this same recent period, so a "record low" quoted without its duration is not automatically comparable to another record low from a different duration class.

Depending on exactly which two of those tender results get compared, the headline "how much have BESS tariffs fallen" number comes out differently — and this is not sloppy reporting so much as a genuinely different question each time. Bookend the opening ₹10.83 lakh pilot against December 2025's ₹1.48 lakh record low and the decline is roughly 86%. Bookend it against the ₹2.22 lakh January 2025 VGF-backed tender instead and it is roughly 80%. Bookend it against the blended ₹3.0–3.2 lakh H2 2024 range and it is roughly 71%. All three percentages — ~71%, ~79.6% and ~86% — turn up in secondary reporting on this market, and none of them is wrong; they are simply answering "decline between which two tenders" differently. This piece is not going to pick one as the correct figure.

What makes that spread matter is the comparison IEEFA and JMK Research draw in their May 2026 joint report: over the same 2022–2025 window in which discovered tariffs fell by however many tens of percent, battery pack (cell) prices fell by only about 36.5%. Tariffs, in other words, have been falling roughly twice as fast as the underlying hardware actually got cheaper. The report's own reading of that gap is not that Indian developers found a genuine cost advantage nobody else has — it is that aggressive, possibly unsustainable bidding is doing the work cost deflation alone can't explain. IEEFA-JMK's model puts a number on the consequence: it estimates that nearly 75% of the two-hour standalone BESS capacity awarded in 2025 falls below its own viability benchmark of ₹2.3 lakh/MW/month. That benchmark is an analyst judgment built into one firm's model, not a government-verified threshold, and should be read as such — but it is a specific, falsifiable claim, not a vague warning.

The knock-on risks analysts attach to that gap are concrete rather than abstract. Lenders reportedly want 15–20% returns on BESS projects, a bar that gets harder to clear the lower a project's own tariff sits. IEEFA frames the market's current phase as an "execution problem" as much as a financing one, citing typical implementation delays of 9–18 months driven by grid-interconnection bottlenecks, financing gaps and procurement delays on top of the margin pressure itself. And the sharpest version of the risk is a safety one: analysts explicitly flag that developers squeezed by aggressive winning bids could compromise battery quality and installation standards to hit their numbers, which raises fire risk and undermines the grid reliability the whole storage build-out is supposed to deliver — the same fire-risk terrain this blog covered from the recycling and battery-passport side in a related piece linked below. One more figure worth flagging as provisional rather than settled: an earlier, separately reported "world's largest renewable-cum-storage firm-supply tender" is said to have cleared at a ₹4.04/kWh (5.61 US cents/kWh) weighted-average tariff fixed for 25 years, but this piece could not independently pin down that tender's name or exact date, so it is presented here as reported, not confirmed.

All figures secondary-press reported on named tenders (see Documents & sources below); duration class and VGF status differ across rows, so these are not a strictly like-for-like series. Battery-pack-cost decline benchmark and viability estimate per IEEFA & JMK Research's joint report on 2025 BESS tender viability (May 2026).

India's Production-Linked Incentive (PLI) Scheme for Advanced Chemistry Cell (ACC) battery storage was approved by the Union Cabinet on 12 May 2021, with an outlay of ₹18,100 crore aimed at 50 GWh of giga-scale ACC manufacturing capacity. As reported, the scheme has awarded 40 GWh of capacity to four beneficiary firms — but only about 1.4 GWh, or 2.8% of the original 50 GWh target, had actually been commissioned as of October 2025, per IEEFA's headline framing of the figure. A later data point, dated 31 May 2026, repeats the same 1.4 GWh commissioned and 40 GWh awarded figures, suggesting little forward movement in actual commissioning across those seven months. In July 2026, the Ministry of Heavy Industries opened a new, 10 GWh tranche of the scheme specifically reserved for Grid-Scale Stationary Storage (GSSS) rather than EV batteries — tender documents released 15 July 2026, a pre-bid conference on 29 July, and a bid-submission deadline of 13 October 2026 — the first ACC PLI tranche explicitly earmarked for grid storage rather than electric vehicles.

The import-dependence picture underneath that gap is stark on the cell side specifically. Multiple 2026 secondary sources converge on the estimate that India's actual lithium-ion cell manufacturing capacity remained only around 1 GWh by the end of 2025, despite roughly 60 GWh of installed domestic battery-pack-assembly capacity — that is, India assembles packs at real scale but still imports almost all of the cells that go inside them. China's estimated share of India's lithium-ion cell imports runs 63–85% or higher, and China is estimated to control 85–98% of global capacity across the upstream components — cathodes, anodes, separators, electrolytes — that feed cell manufacturing everywhere. Those ranges come from trade-analysis outlets (IndexBox, EnkiAI, SilicIndia Energies) rather than an official customs dataset and should be read as informed estimates, not government trade statistics. CareEdge, the ratings agency, is cited secondarily as forecasting that lithium-ion import dependence "may fall to 20% by FY2027" if announced gigafactories come online — with the same source noting those gigafactories may still import the cells themselves, meaning the cell-versus-pack distinction persists even in that more optimistic scenario. Wood Mackenzie, per a separate secondary citation, puts India "5–10 years away" from cell-manufacturing self-sufficiency for battery storage.

One figure this piece is flagging explicitly rather than using as a clean number: the "60 GWh of installed domestic pack-assembly capacity" cited above describes pack-assembly capacity generically, without separating EV-battery-pack lines from grid-BESS-pack lines. The same LFP or NMC cell chemistries, and often the same physical assembly lines, serve both markets in India today, which means grid-storage-specific import dependence cannot currently be isolated from EV-battery import dependence using the public reporting this piece could find. On chemistry more narrowly, one industry-trade source states lithium iron phosphate (LFP) commands roughly 72% of India's BESS chemistry deployments; that specific percentage traces to a single trade blog and could not be cross-checked against a second independent source, so it is presented here with that caveat rather than as settled. Budget 2026-27 removed customs duties on lithium battery cell manufacturing machinery, solar glass inputs and critical-minerals processing equipment — a genuine primary policy move, reached in this piece's research only via pv-magazine India's February 2026 coverage rather than a direct reading of the budget documents themselves.

The single most useful thing this piece can say about India's storage-market data is that it does not currently reconcile with itself, across three separate places, and none of the three gaps below should be silently averaged away.

Cumulative tendered capacity. IESA's own year-end tracker, as reported by pv-magazine India in January 2026, states that 224 GWh of storage capacity had been cumulatively tendered in India by 31 December 2025 — 92 GWh BESS plus 132 GWh PSP, with 95 GWh in various stages of execution, 80 GWh still under tender and 47 GWh of tenders cancelled. A separate, larger figure from an IEEFA-JMK Research joint report states India's cumulative tendered energy storage capacity grew from 6.8 GW in 2018 to 90.7 GW by 2025. These two figures are not directly reconcilable: one is denominated in GWh (energy capacity), the other in GW (power capacity), and press summaries of both trackers mix the two units inconsistently enough that this piece cannot convert one into the other with any confidence. Treat both as directionally indicative of very rapid tendering growth, not as competing answers to the same question.

PSP pipeline stage-counts. Three separate, recent secondary reports citing CEA data give materially different snapshots of India's pumped-storage pipeline. One, citing CEA around June 2026, describes 8 projects (10 GW) under construction, 3 projects (about 3 GW) DPR-concurred, and 49 projects (66 GW) under survey and investigation, with CEA aiming to approve at least 13 PSPs totalling 22 GW in FY2025-26. Another, a Prayas (Pune) piece, gives total PSP potential of about 176 GW, of which 4.7 GW is under operation, 4 GW under construction, 3.6 GW DPR-concurred and about 60 GW under survey. A third, Mercom in March 2026, gives 7.2 GW installed, 5.7 GW operational. "8 projects/10 GW under construction" and "4 GW under construction" do not describe the same pipeline snapshot cleanly, and this piece could not resolve why — plausibly different report dates, or different definitions of "under construction" (civil works begun vs. DPR-approved vs. contract-awarded), but this is an open discrepancy in the public reporting, not something this piece is in a position to settle, particularly since cea.nic.in itself was unreachable in this research.

Government storage-requirement trajectories. At least three primary-government documents, of different vintages, project different storage-requirement numbers, and they should not be merged into one trend line. CEA's Report on Optimal Generation Capacity Mix for 2029-30 (Version 2.0, May 2023) projects a total FY2029-30 storage requirement of 60.63 GW/336.4 GWh (18.98 GW/128.15 GWh PSP plus 41.65 GW/208.25 GWh BESS). The National Electricity Plan 2023, quoted inside the Ministry of Power's National Framework, separately projects 16.13 GW/82.37 GWh for FY2026-27, rising to 73.93 GW/411.4 GWh by FY2031-32. A third, more recent figure — CEA's National Generation Adequacy Plan for FY2026-27 to FY2035-36, cited by Renewable Watch in March 2026 — projects a storage buildout of 174 GW/888 GWh by FY2035-36, including 94 GW of pumped storage. All three trajectories point the same direction — something like a tenfold-plus expansion of required storage capacity over the current decade — but they are three separate modelling exercises, not one number revised over time, and press coverage sometimes cites one, sometimes another, without flagging which.

Layer on top of all three of those the tariff-decline-magnitude spread from Section 3 (~71% vs. ~79.6% vs. ~86%, depending on which tenders are compared) and a pattern emerges that is itself the finding, not a footnote to it: even the reputable trackers in this space — IESA, Mercom, IEEFA-JMK, and secondary press citing CEA — do not agree on cumulative tendered capacity, PSP pipeline stage-counts, or consistent units, let alone a single number for any of the above. There is, as best this piece could determine, no single reconciled public dashboard of India's storage capacity — installed, under construction, tendered, cancelled — that this research could locate or access directly. And every one of the primary-government figures cited throughout this piece — PIB, MNRE, CEA, National Electricity Plan — reached this piece only through search-engine summaries or secondary press citation of those documents; pib.gov.in, mnre.gov.in, cea.nic.in and indiaesa.info all returned blocked-egress errors on every direct-fetch attempt in this piece's research. Read every "per CEA" or "per PIB" line in this piece as primary-by-attribution, secondary-by-access, not as an independently verified government read.