GPU × use-case guide · Embeddings

Is the A100 40GB SXM a good GPU for embeddings?

The A100 40GB SXM is a ampere NVIDIA GPU with 40 GB HBM2e (1555 GB/s), 312 TFLOPS BF16 / 312 TFLOPS FP8, and a 400 W TDP. Embeddings workloads care most about maximum batch throughput on small encoder models — pure tokens-per-dollar. Here's how the A100 40GB SXM measures up.

What models fit on a single A100 40GB SXM?

Weights only, reserving ~25% of the 40 GB for KV cache, activations and fragmentation. ✓ = fits on one card.

Largest single-card fit: Qwen 2.5 14B at BF16, Gemma 2 27B at FP8, Mixtral 8x7B (MoE) at INT4. Bigger models need tensor-parallel across 8 cards.

A100 40GB SXM for embeddings, specifically

Embeddings is throughput-bound rather than VRAM-bound, so the A100 40GB SXM's 1555 GB/s of bandwidth and 312 TFLOPS matter more than raw capacity. Because it runs offline, batch aggressively to push tokens-per-dollar down. Size it precisely on the calculator.

A100 40GB SXM pricing across providers

Verdict

At 40 GB, the A100 40GB SXM is a solid mid-to-high-tier choice for embeddings: single-card up to Qwen 2.5 14B (BF16) or Gemma 2 27B (FP8), and cost-effective at ~$1.19/hr.