Keep lights on
Prioritise battery and backup for high-risk hours instead of wasting them early.
For India's next solar decade
Every society can run like a smart power plant.
Solar has scaled. Now the hard problem is local dispatch: when to store, buy, shed, use backup, or sell flexibility so communities save money and keep the lights on.
Why now
India has already built the solar base. The next layer is intelligence.
MNRE reports 150.26 GW of cumulative solar capacity as of March 31, 2026, including 25.73 GW of grid-connected rooftop solar. PIB's India Solar Momentum note shows how fast this moved: from 3 GW in 2014 to 129.92 GW by October 2025, with nearly 24 lakh PM Surya Ghar households already solarised by December 2025.
150.26 GWsolar capacity reported by MNRE
25.73 GWgrid-connected rooftop solar
24 lakhPM Surya Ghar households noted by PIB
1 crorenational rooftop household ambition
Context sources: MNRE Physical Achievements and PIB India Solar Momentum.
The new operating problem
A rooftop solar site is no longer just a bill reducer.
An apartment society, campus, hospital, market, cold storage unit, EV hub, or rural microgrid is becoming a small energy business. It has solar, battery, grid tariffs, backup fuel, comfort expectations, outage risk, and sometimes export or demand-response opportunity.
Running that well usually needs a capable energy manager. GridOps asks a sharper question: can we turn that expertise into a Capabl Machine?
Apartment societies
RWAs and townships
Schools and campuses
Hospitals and clinics
EV charging hubs
Rural microgrids
What intelligence unlocks
From solar asset to earning, resilient infrastructure.
A model that understands local state can do more than answer questions. It can decide. Charge before a price spike. Save battery before an outage. Use diesel only when it protects critical load. Reduce peak demand. Export or shift load when the economics make sense.
Lower operating cost
Buy less from the grid during expensive windows and avoid unnecessary diesel burn.
Earn from flexibility
Make solar, storage, EV chargers, and flexible loads act like a coordinated local power plant.
How we built the Capabl Machine
Environment first. Model second. Evidence always.
GridOps is an OpenEnv microgrid environment with a stable API. The model sees hourly observations and must output one valid action: {"battery_dispatch": ..., "diesel_dispatch": ..., "demand_shedding": ...}. No vague prose. No fake expertise. The environment executes the action and scores the outcome.
Simulate reality
Demand, solar, price, battery, fuel, heatwave stress, and outage constraints.
Teach the curriculum
1,418 traces across normal, heatwave, crisis, and targeted edge cases.
Train compactly
Qwen2.5-3B-Instruct with QLoRA SFT to learn reliable JSON actions.
Evaluate honestly
Heldout seeds compare do-nothing, trained model, and oracle policy.
What changed after training
The model learned useful battery behaviour, not a shortcut.
Across heldout seeds, the SFT model beat the do-nothing baseline on every task, reached 99.85% valid JSON actions, reduced blackout energy, and used the battery heavily in the hardest crisis setting.
| Policy | Normal | Heatwave | Crisis | Average | What it means |
|---|---|---|---|---|---|
| Do-nothing baseline | 0.5820 | 0.5057 | 0.4522 | 0.5133 | Solar exists, but no intelligent dispatch. |
| SFT action model | 0.6615 | 0.7300 | 0.6648 | 0.6854 | Reliable actions, lower outage impact, better energy use. |
| Oracle policy | 0.7932 | 0.8087 | 0.7046 | 0.7688 | Reference headroom for the next training phase. |
+33%average score lift vs do-nothing
99.85%valid JSON action rate
2,898 kWhcrisis battery throughput
979 kWhcrisis blackout kWh vs 2,426 baseline
The larger point
Capable people will still matter. But every site should not need one full-time.
The opportunity is to package operational judgement into a machine layer that local teams can trust: society managers, solar installers, facility operators, EV hub owners, campus admins, and microgrid developers. That is the Capabl Machines thesis in one working environment.