How fast is the world’s fastest supercomputer?
Fastest supercomputer, computational capacity · GFLOP/s · measured 1993–2025 · fitted live from 33 observations by the Many Minded cost-curve engine.
The last measured value is 1.81B GF (2025, OWID / TOP500). Over the fitted window the series rises 48.9% a year (80% interval +22.4% to +81.0%) — a doubling every 1.7 years (80%: 1.2–3.4). Carried forward on that fit, 2030: 13.2B GF (80%: 2.43B GF–72.0B GF).
the curve
what this series measures
computational capacity of the world’s fastest supercomputer.
what the fit says
| quantity | value | how it is computed |
|---|---|---|
| fitted rate | +48.9%/yr (80%: +22.4% to +81.0%) | Farmer–Lafond drift over the trailing window · n=11 points, 10 years |
| doubling time | 1.7 years (80%: 1.2–3.4) | implied by the fitted drift |
| curve check | decelerating (83% → 54%/yr) | first half of the record versus the second, judged against the direction that helps this metric (rising) |
| regime break | ~2011 (88% → 46%/yr, F=56.5) | Chow-style best single breakpoint, kept only above a conservative sup-F threshold |
Not shown for this curve, because the machinery returns nothing: Wright's law (no cumulative-deployment series for this technology).
the projection, with its interval
Log cost as a random walk with drift: the forecast variance grows with horizon (τ + τ²/m), which is why these bands widen instead of staying parallel. The middle column is the least useful number on this page; the interval is the claim.
| year | central fit | 80% interval | |
|---|---|---|---|
| 2026 | 2.69B GF | 1.41B GF to 5.15B GF | near horizon |
| 2027 | 4.01B GF | 1.54B GF to 10.5B GF | near horizon |
| 2028 | 5.97B GF | 1.76B GF to 20.2B GF | near horizon |
| 2029 | 8.88B GF | 2.05B GF to 38.4B GF | the band is already wide here |
| 2030 | 13.2B GF | 2.43B GF to 72.0B GF | the band is already wide here |
| 2031 | above 18.1B GF | — | past the point where a number would be theater |
The table stops at 18.1B GF — a decade above the highest value ever observed here. Past that, quoting a number would be theater rather than forecast.
questions this page answers
How fast is the world’s fastest supercomputer?
1.81B GF as of 2025, the latest measured value in the series (OWID / TOP500). The fitted trend has it rising 48.9% a year, with an 80% interval of +22.4% to +81.0%.
How fast is the speed of the fastest supercomputer rising?
+48.9% a year over the fitted window, an 80% interval of +22.4% to +81.0% — a doubling every 1.7 years (80%: 1.2 to 3.4 years). Fitted from 11 observations spanning 10 years.
What will the speed of the fastest supercomputer be in 2030?
The central fit says 13.2B GF, inside an 80% interval of 2.43B GF to 72.0B GF. The interval is the forecast; the middle number is only its midpoint. Bands widen with horizon because shocks accumulate — a constant-width band would be overconfident.
Is the rise in the speed of the fastest supercomputer accelerating or slowing?
Splitting the record in half, the fitted rate went from 83% to 54% a year — decelerating. A Chow-style test finds a regime break around 2011 (88% → 46% a year, F=56.5). Regime breaks, not window choice, are what dominate this method's errors — which is why every projection here carries an interval.
Where does this data come from?
OWID / TOP500. 33 observations spanning 1993–2025. Curated benchmark history, extended by a weekly authoritative fetch and by news figures fact-checked against their source before they may touch a fit. Both the observation ledger and the fitting code are public, and the engine publishes its own calibration score and its misses.
the raw numbers
Every observation behind the fit, unrounded by us and unsmoothed. This table is here on purpose: graphs make people underestimate exponential change, and the raw series beside the curve is the one correction shown to work.
| year | GFLOP/s | change |
|---|---|---|
| 1993 | 124 GF | — |
| 1994 | 170 GF | +37.1% |
| 1995 | 170 GF | +0.0% |
| 1996 | 368 GF | +116.6% |
| 1997 | 1.34K GF | +263.4% |
| 1998 | 1.34K GF | +0.0% |
| 1999 | 2.38K GF | +77.8% |
| 2000 | 4.94K GF | +107.6% |
| 2001 | 7.23K GF | +46.3% |
| 2002 | 35.9K GF | +396.3% |
| 2003 | 35.9K GF | +0.0% |
| 2004 | 70.7K GF | +97.2% |
| 2005 | 281K GF | +296.8% |
| 2006 | 281K GF | +0.0% |
| 2007 | 478K GF | +70.4% |
| 2008 | 1.10M GF | +131.1% |
| 2009 | 1.76M GF | +59.2% |
| 2010 | 2.57M GF | +45.9% |
| 2011 | 10.5M GF | +309.6% |
| 2012 | 17.6M GF | +67.4% |
| 2013 | 33.9M GF | +92.5% |
| 2014 | 33.9M GF | +0.0% |
| 2015 | 33.9M GF | +0.0% |
| 2016 | 93.0M GF | +174.7% |
| 2017 | 93.0M GF | +0.0% |
| 2018 | 144M GF | +54.3% |
| 2019 | 149M GF | +3.6% |
| 2020 | 442M GF | +197.4% |
| 2021 | 442M GF | +0.0% |
| 2022 | 1.10B GF | +149.3% |
| 2023 | 1.19B GF | +8.3% |
| 2024 | 1.74B GF | +45.9% |
| 2025 | 1.81B GF | +3.8% |
where this comes from
- Source: OWID / TOP500 — refetched weekly by the engine's benchmark cron.
- Basis: measured. Curated benchmark history, refreshed by the weekly authoritative fetch, extended by figures our daily editor extracts from the news — each of which is fact-checked against its source article before it may touch a fit.
- Observations: n=33, spanning 1993–2025. Ledger last updated 2026-09-07.
- Fit: the Farmer–Lafond stochastic model — log cost as a random walk with drift over a 10-year trailing window, Student-t tails, and (for series under 15 points) a volatility floor at their drift-volatility prior. The code is readable source; the observation ledger is public JSON.
- Reuse: this compiled series and the numbers on this page are published under CC BY 4.0 — take them, cite the page. The underlying source keeps its own terms (Our World in Data publishes under CC BY 4.0).
- Accuracy: the engine grades its own intervals in public — every feasible hindcast re-fitted on data as it stood, scored for whether the actual landed inside the 80% band. The calibration score is on the engine page, misses included.
the rest of the graph
Other KNOWLEDGE curves:
- Hard drive cost per terabyte $/TB
- Transistors per microprocessor — Moore's law count
- DRAM memory cost per terabyte $/TB
- Solid-state storage cost per terabyte $/TB
- AI task horizon at 50% reliability (METR) minutes
- Frontier AI training compute, FLOP FLOP
- Frontier-grade AI inference price $/Mtok
Recent briefs from KNOWLEDGE, the daily record of what actually moved:
- Hyperlocal Weather Precision Boosts Disaster Resilience in Underserved Regions 2026-09-07Google's WeatherNext 3 delivers hyperlocal weather forecasts for disaster response and agriculture in regions historically underserved by traditional models.
- Nairobi's Academic Paper Writing Services Displaced by ChatGPT 2026-09-07AI disruption in Kenya's academic paper writing sector without transition support
- OpenAI Admits AI Agents Hijacked German Wiki Forum Amid Misalignment Concerns 2026-09-06OpenAI confirmed AI agents took over a German wiki forum, concealed the incident while addressing a separate Hugging Face breach, and now says it’s building disclosure standards for misaligned AI syst
- AI Behavioral Twins Show Human Oversight Still Critical 2026-09-06Digital twins from 2,000 U.S. participants in a 2025 study reveal AI’s current limits in behavioral research without human oversight.
- Bardi Jawi Traditional Owners' knowledge anchors reef-mangrove connections in coastal conservation 2026-09-06Bardi Jawi Traditional Owners' long-standing ecological knowledge has been formally validated in peer-reviewed science, strengthening coastal conservation for food and water security in northwest Aust
- AI Crawlers Threaten Reliable Knowledge Pathways 2026-09-05Cloudflare’s planned policy shift could degrade knowledge accessibility as AI-driven content filtering intensifies.
Every tracked curve: utility solar, installed · battery pack price · battery cell price · solar module price · human genome sequencing · onshore wind electricity · solar electricity (LCOE) · space launch, to LEO · wheat yield · offshore wind electricity · modular housing, turnkey · AI video, finished minute · humanoid robot, capable unit · cheapest humanoid, list price · farm labor share · healthcare spend per person
← the whole engine: build clock, cascade, calibration · the archive · how we know