Startup Fundraising

Stepfun Raises $2.5B for On-Device AI

Chinese AI startup Stepfun secures $2.5 billion from Huaqin Technology, ZTE, and others to advance its on-device AI strategy for smartphones and automotive.

Share:
AM
Alvaro de la Maza

Partner at Aninver

Stay ahead of the market

Get instant notifications when new news matching "Artificial Intelligence (AI), Technology, Software & Gaming in China" are published.

Key Takeaways

  • 階躍星辰 (Stepfun) raised $2.5B (Series A) from 華勤技術 (Huaqin Technology), 中興通訊 (ZTE).
  • Sector: Artificial Intelligence (AI), Technology, Software & Gaming.
  • Geography: China.

Analysis

Chinese AI innovator Stepfun is reportedly in advanced discussions to secure approximately $2.5 billion (around 400 billion JPY) in a significant funding round. This substantial capital injection is expected to fuel the company's ambitious strategy to embed advanced AI models directly into consumer electronics and other end-user devices. The round is notably characterized by strong industrial backing, with key participation from hardware manufacturers like Huaqin Technology and telecommunications giant ZTE, signaling a deep commitment from the hardware ecosystem to on-device AI capabilities.

Founded in April 2023, Stepfun has rapidly established itself in the competitive AI model development space. Its proprietary 'Step' series of large language models (LLMs) demonstrate impressive multimodal performance, spanning text, voice, image, and video processing, alongside complex reasoning tasks. The latest iteration, 'Step 3', is recognized for its advanced visual recognition and sophisticated inference engines, positioning the company at the forefront of AI model innovation in China. This funding will be crucial for scaling these advanced models for widespread deployment.

Stepfun's core strategy revolves around creating 'super assistants' by integrating its AI models into a variety of hardware platforms. The company is actively collaborating with leading smartphone manufacturers such as OPPO and Honor, with its AI already pre-installed on over 42 million devices. Beyond mobile, Stepfun is making significant inroads into the automotive sector, partnering with Geely Automobile to deploy its 'AgentOS' smart cabin solution. This initiative aims to transform vehicle interiors into intelligent, interactive environments.

Further solidifying its position, Stepfun has forged a strategic alliance with Tencent Cloud. This partnership focuses on the joint development of AI cockpit agents for vehicles, promising a more integrated and seamless user experience within cars. The collaboration seeks to unify services like music streaming, video playback, navigation, and payment systems, creating a comprehensive in-car digital ecosystem powered by advanced AI. The influx of capital will accelerate these development efforts and expand their reach.

This significant funding round for Stepfun follows a trend of substantial investments in China's AI sector, with companies like DeepSeek and Moonshot AI also attracting considerable capital. The focus on on-device AI represents a critical shift, moving powerful AI capabilities from the cloud directly to the user's hardware. This approach promises enhanced privacy, reduced latency, and more personalized user experiences, aligning with the growing demand for intelligent, responsive devices across the consumer electronics and automotive industries. The involvement of major hardware players underscores the industry's belief in the future of edge AI.

The strategic alignment with hardware manufacturers like Huaqin Technology and ZTE is particularly noteworthy. These companies are integral to the production of smartphones and consumer electronics, providing Stepfun with direct channels to integrate its AI solutions into mass-market products. This synergy is expected to accelerate the adoption of advanced AI features, potentially setting new benchmarks for user interaction and device intelligence in the consumer technology space. The substantial investment underscores the immense market potential anticipated for AI-powered edge computing.