Quanfluence Raised $10 Million To Build Four Qubits, And Sells Something Else Meanwhile

In this storyQuanfluence

Quanfluence has raised $10 million in a round led by Chiratae Ventures, with participation from Rainmatter by Zerodha and existing investor Pi Ventures.

The Bengaluru company had previously raised $2 million in a seed round led by Pi Ventures in December 2024, with Golden Sparrow, Reena Dayal and others participating.

The money will fund a four-qubit quantum computer in the near term, with a prototype of around 100 qubits targeted by 2029 and a fault-tolerant, general-purpose machine over five to six years.

Founded in 2021 by Sujoy Chakravarty, Ravi Mehta, Biman Chattopadhyay, Anil Prabhakar, Aditi Vaidya and Sandeep Goyal, Quanfluence builds photonic quantum computing hardware for optimisation problems, positioned as a complement to AI in decision-making.

Its optical Ising machine combines light waves with FPGAs for optimisation processing, which the company says can run up to 100 times faster than classical computing on certain workloads. Unlike superconducting systems that require complex cryogenic cooling, the technology operates at room temperature. The company is testing a version capable of processing 10 to 25 million permutations, aimed at financial risk management, logistics and delivery route planning, alongside workforce scheduling.

The sector sits within the government's ₹6,003.65 crore National Quantum Mission, which includes photonics among the platforms it is developing.

Other Indian companies in the field have raised recently. QpiAI took $32 million in Series A from Avataar Ventures and the National Quantum Mission last year, and borrowed ₹50 crore from InnoVen Capital this month. QuBeats raised $15 million in April from Zoho Corporation, Indusbridge Ventures and the National Quantum Mission. BQP, formerly BosonQ Psi, raised about $5 million this year for quantum-inspired simulation.

Four qubits is a very small number, and the gap between it and a hundred by 2029 is where the interesting part of this company sits.

A four-qubit machine computes nothing commercially useful. Neither does a hundred, for most purposes. Fault tolerance, which is the point at which quantum computers do things classical ones cannot, needs orders of magnitude more. Quanfluence puts that five to six years out, which is the standard answer across the industry and has been for a decade.

What makes the round legible is that the roadmap is not the business. The optical Ising machine already works and solves a real category of problem: high-dimensional optimisation, the kind that appears in delivery routing, shift rostering and portfolio risk. It uses quantum-adjacent physics rather than qubits, runs at room temperature and plugs into ordinary infrastructure.

That is a product with customers today, funding a research programme that may pay off much later. It is the same structure that let Nvidia spend years on CUDA while selling graphics cards, and it is a considerably better position than a company whose only asset is a qubit count on a slide.

The risk is the reverse of the usual one. Not that the science fails, but that the near-term optimisation business turns out to be a modest niche against well-tuned classical solvers, leaving a long-dated research effort without a cash engine underneath it.

India now has four distinct quantum approaches funded, and they are not competing for the same thing.

QpiAI is building superconducting qubits and full-stack systems, and has just borrowed ₹50 crore at 13.85 per cent against a 1,000-qubit target for 2028. QuBeats works on quantum sensing and lasers. BQP sells quantum-inspired simulation that runs on classical hardware. Quanfluence does optical optimisation at room temperature.

Only one of those requires the general-purpose machine to arrive for the business to work. The National Quantum Mission has backed three of the four directly, which looks less like picking a winner than like buying options across incompatible technical bets, which is the sensible way to fund a field where nobody knows which platform wins.

The room temperature point is the commercial difference. A superconducting system needs a dilution refrigerator, a specialist facility and staff who can run both. An optical machine that works at ambient conditions can sit in an ordinary data centre, which means it can be sold to a bank or a logistics operator rather than only to a research institution. That changes the addressable customer set entirely.

What is not yet visible is revenue. The release describes testing and targets rather than deployments, and optimisation is a crowded field where classical solvers have improved considerably. The claim of up to 100 times faster on certain workloads is the company's, and the qualifier is doing real work in that sentence.

The honest read is a well-funded company with a plausible near-term product and a long-term ambition, where the first will pay for the second if customers materialise.

For daily, sharp analysis of the biggest moves in the Indian business and startup ecosystem, follow StartupFox.