TL;DR
FirstQFM reported that its quantum reservoir computing system outperformed the classical models tested on next-day forecasts across 43 anonymized assets and 600 out-of-sample periods. The company reported 52.02% directional accuracy and 112% higher cumulative gross portfolio profit than the strongest classical benchmark by profit; the findings come from a company-announced evaluation, and independent replication is not established in the supplied material.
FirstQFM said November 3 that its quantum reservoir computing system, with quantum features generated on Rigetti Computing hardware, outperformed the classical forecasting models tested in an evaluation spanning 43 anonymized financial assets and 600 out-of-sample periods. The company reported 52.02% directional accuracy and 112% higher cumulative gross portfolio profit than the strongest classical model by that measure, but the results are a company-announced evaluation rather than evidence of independently replicated performance.
Get wellness gear delivered free with Prime
- Fast, free delivery on millions of items
- Prime Video, Amazon Music and more included
- Member-only deals all year
The evaluation covered next-day forecasts and compared FirstQFM’s foundation-model-based quantum reservoir computing (QRC) system with statistical and machine-learning models, as well as time-series foundation models. The named examples included Chronos-2 from Amazon, Granite TTM-R2 from IBM and Toto 2.0 from Datadog. FirstQFM said its system recorded the highest directional accuracy and macro-F1 among the models evaluated, with a macro-F1 score of 51.76%.
For an equally weighted portfolio of all 43 assets, the company reported cumulative gross profit and loss (P&L) 112% higher than that of the strongest classical benchmark by P&L. Its annualized Sharpe ratio was 1.22, compared with 1.02 for the classical model with the highest Sharpe ratio. FirstQFM also said the system had directional accuracy above 50% on 36 of 43 assets and positive gross P&L on 41 of 43.
The data and forecasting task were anonymized by SW7 Group before being supplied to FirstQFM. The report says the test used chronological training, validation and test periods, and model selection was based only on validation data. FirstQFM also described a separate five-asset portfolio selected using a historical diversification method: it reported better directional accuracy, macro-F1 and cumulative gross P&L than the 43-asset portfolio, with a 1.73 annualized Sharpe ratio.
What the Forecasting Results Could Mean
The announcement adds a finance-focused performance result to continuing efforts to test whether quantum computing can contribute to practical machine-learning tasks. The reported measures matter because forecast accuracy and portfolio risk-adjusted returns address different questions: directional accuracy tracks whether a prediction gets movement direction right, while the Sharpe ratio compares return with variability. FirstQFM reported an advantage on both in this evaluation.
For investors and financial firms, however, the reported figures do not establish that the system will produce similar results in live trading. The comparison concerns gross P&L, not net returns after transaction costs, fees or market impact, and the source material does not establish independent verification. The results are relevant as an experiment, but they do not by themselves show that a quantum system is commercially superior across markets or over time.
Products Worth Considering
DEWALT 20V MAX Cordless Drill and Impact Driver Combo Kit (DCK277D2)
WORX WX838L.1 20V Cordless Brushless 18GA Brad Nailer
EGO POWER+ 16-Inch Cordless String Trimmer with 4.0Ah Battery and Charger (ST1623T)
As an affiliate, we earn on qualifying purchases.
How FirstQFM Ran the Test
FirstQFM describes its approach as foundation-model-based quantum reservoir computing, with quantum features generated using Rigetti’s Cepheus-1-108Q system. The evaluation used anonymized financial data from SW7 Group and compared forecasts against classical alternatives, including established model types and named time-series systems. The chronological split and validation-only model selection, as described in the announcement, are intended to keep the test period separate from the model-selection process.
FirstQFM released a beta version of its forecasting system in July 2026, according to the supplied report. The company says its underlying foundation-model technology is built at billion-parameter scale and is intended for multiple sectors. Those broader commercial and cross-industry plans are company descriptions; the results announced here concern a particular financial forecasting evaluation, not performance across all those applications.
“Empirical work such as FirstQFM’s helps us probe the capabilities and limits of today’s quantum computers, while helping hardware teams understand which improvements matter most.”
— Subodh Kulkarni, Rigetti Computing chief executive
Limits of the Reported Comparison
The supplied material does not provide an independent replication, full test data or enough detail to assess how the reported results would change under different assets, time periods or trading assumptions. Although it describes the data split and model-selection process, the detailed technical methodology is not reproduced in the announcement; FirstQFM points readers to an accompanying technical blog for more information.
It is also unclear from the supplied information whether the gross portfolio results include realistic transaction costs, fees, slippage or market impact, and no live-trading results are reported. The 112% figure is a comparison with the strongest classical benchmark by cumulative gross P&L in this evaluation; it should not be read as a return for investors or as a general performance guarantee. The source does not state the absolute P&L amounts or the test’s calendar dates.
Technical Details and Further Testing
FirstQFM has directed readers to its technical blog for the detailed methodology and results. Further scrutiny would depend on the information it provides there, including the evaluation setup, portfolio calculations and any additional testing. The supplied announcement does not give a timetable for independent replication or a live deployment, so those remain open questions.
For now, the next meaningful evidence would be results that clarify performance after trading costs and show whether the reported advantage persists across additional periods and datasets. Until then, the announcement stands as a reported benchmark comparison from FirstQFM’s test, not a settled finding about quantum forecasting in financial markets.
Key Questions
What did FirstQFM report?
FirstQFM said its quantum reservoir computing system outperformed the classical models tested on next-day forecasts across 43 anonymized assets and 600 out-of-sample periods. It reported 52.02% directional accuracy and 112% higher cumulative gross P&L than the strongest classical benchmark by P&L.
Which quantum hardware was used?
The announcement says the system’s quantum features were generated using Rigetti Computing’s Cepheus-1-108Q system.
Were the results independently verified?
The supplied material describes results announced by FirstQFM and does not establish independent verification or replication. The company points to a technical blog for its detailed methodology and results.
Does the reported profit figure represent investor returns?
No. FirstQFM reported cumulative gross portfolio P&L for an equally weighted test portfolio. The supplied information does not say whether transaction costs, fees or market impact were included, and it does not report live-trading returns.
Source: thequantuminsider.com
