GPU × use-case guide · Translation

Is the A10G a good GPU for translation?

The A10G is a ampere NVIDIA GPU with 24 GB GDDR6 (600 GB/s), 35 TFLOPS BF16 / 35 TFLOPS FP8, and a 150 W TDP. Translation workloads care most about high batch throughput and cost efficiency across many short-to-medium requests. Here's how the A10G measures up.

What models fit on a single A10G?

Weights only, reserving ~25% of the 24 GB for KV cache, activations and fragmentation. ✓ = fits on one card.

Largest single-card fit: Llama 3.1 8B at BF16, Qwen 2.5 14B at FP8, Gemma 2 27B at INT4. Bigger models need tensor-parallel across 8 cards.

A10G for translation, specifically

Translation is throughput-bound rather than VRAM-bound, so the A10G's 600 GB/s of bandwidth and 35 TFLOPS matter more than raw capacity. Because it runs offline, batch aggressively to push tokens-per-dollar down. Size it precisely on the calculator.

A10G pricing across providers

Verdict

At 24 GB, the A10G is a value card best suited to smaller translation models (up to Gemma 2 27B with INT4 quantisation). For larger models you'll want more VRAM or multi-GPU.