The archive · Hardware & Devices · Technical decision · 2025–2026
SynaXG bets AI-native radio networks become physical AI's foundation
Singapore deep-tech startup SynaXG raises over US$20M in its first round to sell AI-native radio access network software for the physical AI era.
SynaXG Technologies
What the business is
SynaXG builds AI-native radio access network (AI-RAN) software — full-stack L1/L2/L3 RAN, virtual DU/CU and radio units tuned for GPU, CPU, FPGA and ASIC compute — for telecom operators and wireless equipment makers.
Starting capital：Over US$20M in its first funding round (December 2025), led by January Capital Growth Credit, Vertex Ventures and Qualgro
How it started
Founded in Singapore by wireless industry veteran Xin Huang with a team carrying decades of RAN expertise, SynaXG says its core group spent nearly four years building and refining AI-RAN technology before disclosing its first funding round.
What happened
In December 2025 SynaXG closed a first funding round of over US$20 million, led by January Capital Growth Credit, Vertex Ventures and Qualgro and described as one of Asia's most significant early-stage AI-RAN investments. Its customers and partners include leading AI-RAN chip companies, network equipment providers and operators, and the company said the capital would accelerate its product roadmap and global engineering teams. The announcement landed weeks after Nvidia revealed a US$1 billion investment in Nokia for AI-RAN development, which SynaXG cites as validation of the category.
No ending yet — it is still running.
Background
SynaXG is a Singapore deep-tech startup betting that AI-RAN — artificial intelligence embedded directly into radio access networks — becomes the core infrastructure of the physical AI era, in which robots, autonomous systems, drones and industrial automation need ultra-low-latency, high-reliability, real-time wireless. Founded by wireless industry veteran Xin Huang, the company sells full-stack RAN software (L1/L2/L3), virtual DU/CU and radio units optimized for GPU, CPU, FPGA and ASIC compute, and says its team spent nearly four years building the technology.
The bet is architectural: instead of today's hardware-centric radio networks, SynaXG argues AI belongs inside the network itself, comparing AI-RAN to 'the iPhone reshaping the mobile era.' Its first external validation came in December 2025, when it closed a first round of over US$20 million led by January Capital Growth Credit, Vertex Ventures and Qualgro — described as one of Asia's most significant early-stage AI-RAN investments.
Customers and partners today include leading AI-RAN chip companies, network equipment providers and operators. SynaXG says the capital will accelerate its product roadmap, strengthen global engineering teams and deepen operator and enterprise collaborations, and it is preparing a Series A aimed at expanding commercial rollouts. The company points to Nvidia's US$1 billion investment in Nokia for AI-RAN development as proof the category is real — a tailwind, but not yet evidence that telecoms will buy its software at scale.
What has to be true
- AI-RAN is a greenfield bet: if physical AI really needs a new wireless layer, the winners are whoever owns the network software, not just silicon.
- January Capital, Vertex Ventures and Qualgro staked one of Asia's most significant early-stage AI-RAN rounds on the thesis.
- Nvidia's US$1B investment in Nokia gives the category external validation from the biggest AI hardware player.
- The open question is demand: SynaXG's public customers today are mostly partners and chip or equipment vendors, and operator purchases at commercial scale are unproven.
What can be applied
Category tailwinds are not demand: SynaXG's round rides Nvidia's US$1B Nokia investment, but whether operators actually buy AI-native RAN software is still untested.
Aftermath
As of September 2026 SynaXG remains pre-Series A: the December 2025 round funds engineering teams, the product roadmap and operator collaborations, and the company has said it is preparing a Series A to expand commercial rollouts and AI-native RAN adoption across global markets. No revenue, operator contract or deployment figures have been disclosed, so the core bet — that telecom operators will buy AI-native RAN software instead of traditional networks — is still unproven.
Sources
- SynaXG Raises Over US$20 Million to Accelerate Global AI-RAN Leadership
- Singapore's SynaXG raises over $20M funding to accelerate global AI-RAN
- SynaXG raises over US$20 million in first funding round
spotted an error? The archive wants to know.
Your turn
You just read one. Describe what you are building, and see who is betting on the same thing.
Free account · 3 free questions · no card