Intel Arc Pro A60 12GB vs Intel Arc Pro A40 6GB

The Arc Pro A60 is the newer part: Alchemist, launched in 2023, against the Arc Pro A40's Alchemist from 2022. Newer architectures typically add lower-precision formats and better throughput per watt, so check the precision rows your workload actually uses.

Dense throughput for the Arc Pro A60 against the Arc Pro A40: FP32 10 vs 5 TFLOPS. Sparse figures, where the vendor publishes them, appear under each dense number in the performance table.

Memory is 12 GB against 6 GB, fed at 384 GB/s versus 192 GB/s. More memory per GPU means larger models fit before you have to shard across cards, which often matters more than raw TFLOPS for inference.

Power budgets are 130 W for the Arc Pro A60 and 50 W for the Arc Pro A40. At FP32 that works out to 0.08 against 0.10 TFLOPS per watt.

For LLM training and inference, weight the FP8 and FP16 rows and memory capacity most heavily. For scientific and HPC workloads, the FP64 row is the one to read.