Hardware Corner Mar-2026: nvidia-rtx-3090 llama.cpp Gemma4 31B Q4_K GGUF prompt processing 723.7 tok/s at 32k context

THIRD-PARTY MEASURED (hardware-corner.net /gpu-llm-benchmarks/rtx-3090/, page updated March 2026, accessed 2026-10-16). llama.cpp llama-bench, GGUF, single GPU, full offload. The page publishes prompt-processing tok/s and token-generation tok/s by context tier; it does not state the llama.cpp build, the prompt length, or the generation length. The page does not annotate flash attention for this model row. Value is the 32k context column. Per-request batch-1, NOT aggregate across concurrent requests.

Context tier is the page bucket, not an exact measured token count. Page quant Q4_K maps to gguf-q4-k-m; page MXFP4 maps to gguf-mxfp4-moe. Distinct from the older Q4_K_XL rows sourced from hardware-corner-gpu-ranking-local-llm.