llama.cpp CUDA RTX 5090 llama-2-7b Q4_0 pp512 FA
14970.15 tok/s (pp512)
Fields
id
benchmarks/llamacpp-scoreboard-rtx-5090-fa-pp512Metric
prefill_tok_s
Value
14,970.15
Unit
tok/s (pp512)
Model
llama-2-7b.Q4_0.gguf (ggml-org, 3.56 GiB, 6.74B params)
Methodology
COMMUNITY-SCOREBOARD (llama.cpp discussion #15013, maintainer-curated CUDA backend). llama-bench llama-2-7b.Q4_0.gguf (ggml-org, 3.56 GiB, 6.74B params), pp512 and tg128, n_gpu_layers 99, batch 1, full GPU offload, CUDA. Per-request batch-1, not an aggregate across concurrent requests. Flash attention on (-fa 1). Build 8cf6b42, reporter @totaldev. Printed mean 14970.15. Re-read of the discussion page on 2026-10-17.
Measured by
third_party
Reproducible
yes
notes
RTX 5090, 32 GB / GDDR7 / 512 bit, build 8cf6b42, reporter @totaldev. prompt processing cell.
Notes
RTX 5090, 32 GB / GDDR7 / 512 bit, build 8cf6b42, reporter @totaldev. prompt processing cell.