ByteDance Eyes Iluvatar CoreX and Baidu AI Chips in Major Shift from Nvidia

ByteDance is negotiating with Iluvatar CoreX and Baidu for AI inference chips, a move that highlights China's accelerating shift to domestic GPU suppliers. The deals could reshape the competitive landscape as Nvidia's presence in China dwindles.

Last Updated: July 28, 2026 Editorial Process
Editorial Process
See more of Inside AI's trusted news by adding us as a preferred source on Google.
AI neural network visualization
By Sophia Andreou Published on: June 15, 2026

June 15, 2026, (Inside AI) — ByteDance, the parent company of TikTok, is in active talks with Shanghai-based chip startup Iluvatar CoreX to acquire AI processors for inference workloads. Two sources with direct knowledge of the discussions confirmed the negotiations, which could make Iluvatar CoreX ByteDance's third major domestic GPU supplier after Huawei and Cambricon.

ByteDance is also exploring a parallel deal with Baidu to use its Kunlunxin chips, the sources said. Tencent already counts as a Kunlunxin customer. None of the companies responded to requests for comment.

The talks underscore a rapid shift in China's AI hardware landscape. Chinese GPU makers captured nearly 41% of the domestic AI accelerator server market last year, eroding Nvidia's once-dominant position. U.S. export controls on advanced chips have pushed Beijing to champion local alternatives, and these potential deals signal that effort is gaining commercial traction.

A Shifting Chip Landscape

Nvidia's market share in China has effectively dropped to zero, CEO Jensen Huang acknowledged. Yet Tencent's Chief Strategy Officer James Mitchell predicted in May that Chinese AI chips would become available in large quantities in the second half of this year. The ByteDance talks appear to validate that forecast.

Iluvatar CoreX expects to ship at least 50,000 chips to ByteDance this year, mostly for inference tasks tied to Doubao, ByteDance's flagship AI chatbot. Inference workloads handle real-time user queries, unlike training which demands cutting-edge compute. This distinction allows Chinese firms to carve out a niche even without access to the most advanced fabrication.

Iluvatar's Commercial Breakout

A deal with ByteDance would mark a commercial milestone for Iluvatar CoreX. Until now, the company relied heavily on government procurement projects, one source noted. Its Hong Kong listing in January followed 1 billion yuan ($148 million) in 2025 revenue, with roughly 90% coming from GPU sales.

The firm's Tiangai series targets AI training, while Zhikai chips focus on inference. Huatai Securities projects revenue will hit 3.04 billion yuan ($449.8 million) this year, with shipments surging 139% to over 100,000 units. The broker estimates Zhikai inference chips sell for about 12,000 yuan ($1,775) each.

Baidu's Quiet Chip Push

ByteDance's interest in Baidu's Kunlunxin chips adds another layer. Baidu has quietly built a semiconductor arm, and Tencent's existing adoption suggests the technology is maturing. The sources cautioned that details remain fluid and no final agreement has been signed.

The broader narrative is clear: China's AI infrastructure is decoupling from U.S. suppliers faster than many expected. While Nvidia still leads globally in training chips, the inference market is proving more accessible for domestic players. ByteDance's sheer scale as an AI spender could accelerate that trend.

For Iluvatar CoreX, the ByteDance talks represent both validation and pressure. Scaling from government contracts to hyperscale commercial deals demands reliability and volume that few Chinese startups have demonstrated. The coming months will test whether the supply chain can deliver.

More from Inside AI

  • AI In Business

    AI Push Is Putting Banks at Mercy of Tech Firms, Warns Moody’s

    August 9, 2026
  • AI Safety

    Anthropic Is Destroying Books to Make AI Training Data

    August 9, 2026
  • AI In Business

    Meta CTO Says AI Productivity Gains Should Mean More Work, Not Extra Time Off

    August 9, 2026
  • AI In Business

    Why Ex-HCL CEO Vineet Nayar Feels the Real AI Risk Isn’t Smarter Machines

    August 9, 2026
  • AI Safety

    OpenAI to Pause Work on AI Model Astra Due to Security Concerns

    August 8, 2026
  • AI In Business

    Apple Mac Users in China Can Connect to Alibaba’s Qwen AI Service

    August 8, 2026
  • Artificial Intelligence (AI), AI Safety

    Parenting is hard. Should we let AI do it for us?

    August 8, 2026
  • AI In Business

    66% of India’s AI Workers Expect Layoffs, Blind Survey Finds

    August 8, 2026

Never Miss a Breakthrough

Join 50,000+ readers who get our daily AI intelligence briefing. No fluff, just what matters.

Inside AI is an independent publication covering artificial intelligence news, machine learning research, and the tools shaping the future of technology. No hype. Just what's happening in the AI world.

Topics

  • Artificial Intelligence
  • Machine Learning
  • Generative AI
  • Agentic AI
  • Vibe Coding
  • Prompt Engineering
  • AI Policy & Regulation
  • AI Hardware & Infrastructure
  • AI Tools
  • AI In Business
  • Robotics
  • Cybersecurity AI
  • AI Safety
  • AI Tools & Reviews (Coming soon)

Company

  • Editorial Standards
  • Privacy Policy
  • Terms of Service
  • Contact
  • About Us

Others

  • Press Releases
  • Features
  • Sponsored Content

© 2026 Inside AI. All rights reserved.

Designed by Blue Flare Digital