RTX PRO 5000 Blackwell Ollama: Qwen2.5 32B eval rate (Database Mart)
42.92 tok/s (eval rate, per-request)
Fields
id
benchmarks/dm-pro5000-ollama-qwen25-32b-evalMetric
decode_tok_s
Value
42.92
Unit
tok/s (eval rate, per-request)
Model
qwen2.5:32b
Methodology
Database Mart hosted GPU-server benchmark of one NVIDIA RTX Pro 5000 Blackwell Server Edition (48 GB GDDR7 ECC, 300 W, PCIe 5.0 x16, driver 580.65.06, CUDA 13.0). Engine is Ollama 0.12.3 via the Ollama API (llama.cpp backend; Ollama build commit not printed). Quantization is the default Ollama pull for each tag, which the report labels INT4 and does not further specify (not asserted as Q4_K_M). Prompt is fixed at 2048 tokens (fixed-length text file, 50 repetitions); output length is not stated. The published column is eval rate (tokens/s), which is decode, per request, batch size 1, NOT an aggregate across concurrent requests. prompt eval rate is a separate column and is not recorded here. Card is the 48 GB Server Edition, not the 72 GB workstation SKU.
Measured by
third_party
Reproducible
yes
notes
Table row qwen2.5:32b, 19 GB model, load 12.03 s, prompt eval 1937.13 tok/s, eval 42.92 tok/s. Computingforgeeks independently measured 42.9 tok/s eval on the same Ollama tag and card class.
Notes
Table row qwen2.5:32b, 19 GB model, load 12.03 s, prompt eval 1937.13 tok/s, eval 42.92 tok/s. Computingforgeeks independently measured 42.9 tok/s eval on the same Ollama tag and card class.