The chips are in the design phase and will enter mass production in a year. One is for AI training, and the other is for AI inference. Broadcom, which has already built AI chips for Google, is expected to design the AI chips.

The GPUs are said to be produced on one of TSMC's N4/N5 process technologies, so a similar node to TSMC's 4NP is used to build Nvidia's Blackwell series GPUs for AI and HPC. ByteDance's GPUs are expected to enter mass production by 2026, so expect their deployment in 2026.

ByteDance has spent more than $2 billion on more than 200,000 Nvidia H20 GPUs (i.e., about $10,000 per unit, which is a bit less than $12,000 – $13,000) for its AI efforts this year, and many of these GPUs have yet to be delivered to the company.

The shortage of Nvidia GPUs and their high price are why ByteDance built its own AI hardware. Nvidia designed its DGX H20 and some other GPUs specifically for the Chinese market in response to U.S. export controls imposed last year.

The HGX H20 is a massively reduced GPU (compared to the H100) that still sells for $10,000. Nvidia's HGX H20 only offers 296 INT8/FP8 TOPS/TFLOPS and 148 BF16/FP16 TFLOPS performance for AI computations, while the fully-fledged H100 delivers 3,958 INT8/FP8 TOPS/TFLOPS as well as 1,979 BF16/FP16 TFLOPS performance for AI computations.

However, with its 96 GB of HBM3 memory, up to 4.0 TB/s of memory bandwidth, and 8-way GPU capability, Nvidia's HGX H20 is still highly demanded by Chinese companies in real-world applications.

While ByteDance will unlikely be able to make its GPUs significantly faster than Nvidia's HGX H20 due to U.S. export control rules (as TSMC will be unable to ship such GPUs), they will be considerably cheaper for the company.