Trainium1 vs Trainium2: specs and rental prices

In short

• The Trainium2 lists 96 GB of GPU memory against 32 GB for the Trainium1: 3× higher.
• Memory type is HBM on the Trainium1 and HBM3 on the Trainium2.
• The Trainium2 lists 2,875 GB/s (derived) of memory bandwidth against 612.5 GB/s (derived) for the Trainium1: 4.69× higher.
• Not every figure can be compared: at least one vendor page does not publish: FP8 tensor throughput; FP16 / BF16 tensor throughput; TF32 tensor throughput; FP64 throughput; Maximum power; GPU-to-GPU interconnect.
• Only the Trainium1 has a fresh on-demand listing ($1.34 per chip-hour at AWS); the Trainium2 has none right now.

This summary is generated from the tables below. Prices were last checked Oct 29, 2026. A listing not re-checked for 7 days is greyed and left out of every lowest-price figure; after 14 days it is removed from the page.

Specifications side by side

Compared versions: Trainium1 Trainium (Trn1) and Trainium2 (Trn2). The figure in the stronger cell is shaded; power is shown without a winner. Two throughput figures are compared only when both vendors state them on the same basis (both with sparsity, or both dense), and “derived” means a total the vendor publishes divided by the GPU count it publishes. Full tables with sources: Trainium1 specifications · Trainium2 specifications.

Rental prices

Prices cover every version and instance size the providers list, so they are not tied to the versions compared above; the price per GB of memory divides the lowest fresh on-demand price by the memory of the compared version. Listed prices are not like for like across providers. See every listing: Trainium1 prices · Trainium2 prices.

More on these GPUs

• AWS Trainium (Trn1): specs, VRAM and rental prices
• AWS Trainium2 (Trn2): specs, VRAM and rental prices
• Trainium2 vs H100
• Trainium2 vs Trainium3
• Inferentia2 vs Trainium1
• All GPU comparisons · Rental cost calculator