The claim going around trade and industry media is that Australia's National Electricity Market, the grid covering the eastern and southern states, hit "record highs" for utility-scale solar generation in September 2026. It's the kind of headline that gets shared as proof the grid's renewable build-out is smashing through milestones every month. The real figures, out of analysis by Rystad Energy and independently crunched data from Open Electricity (formerly OpenNEM), tell a more specific and slightly less tidy story.

What's genuinely true. New South Wales, Queensland and Victoria each posted their best-ever September for utility-scale solar generation, according to Rystad Energy analyst David Dixon's figures as reported by pv magazine Australia and pv-tech.org. NSW generated 841GWh, Queensland 645GWh and Victoria 267GWh from utility PV. Combined with wind, the NEM's utility-scale solar and wind output reached 5.44TWh for the month, up 7% on September 2025. Curtailment, the share of available solar generation the grid couldn't actually use and had to throw away, also fell: from 26% a year earlier to 22% this September. That's a real, measurable improvement in how much of the sunshine hitting those panels is making it onto the grid.

Where the "record" framing overreaches. Reporting the state-level figures as a NEM-wide "record high" implies a national, all-time solar record. It isn't one. A separate analysis of Open Electricity's own data, also published by pv-tech.org, puts national utility-scale solar generation at 1,805GWh for September, which the article calls the strongest September in its three-year dataset, not an all-time high. By that same dataset, the national monthly record for utility solar was actually set in February 2026, at 1,836GWh, which September didn't beat. The widely quoted 5.44TWh figure doesn't help here either: it's solar and wind added together, not a solar number at all, so using it to back up a "solar record" headline is mixing two different fuels into one.

A smaller wrinkle. Victoria's utility solar figure for September also appears slightly differently depending on which write-up you read: 267GWh in pv-tech.org's coverage, 276GWh elsewhere citing the same Rystad analysis. It's a small gap and doesn't change the overall picture, but it's a reminder that these are secondhand industry estimates rather than numbers AEMO itself has published and badged as official.

The bottom line: this one's partly true. The underlying growth is real and worth reporting, three states did have their best Septembers ever for utility solar, and less power is being wasted to curtailment than a year ago. But "record highs in the NEM" oversells it: there's no evidence of a national all-time solar record this September, and the headline number people are sharing around measures solar and wind together, not solar on its own.