NVIDIA RTX 6000 Ada llama.cpp: Llama 3.3 70B (Q4_K) prompt processing at 16k context
526.0 tok/s (pp16k, per-request batch-1)
Fields
id
benchmarks/hc26-6000ada-llama33-70b-q4k-16k-ppMetric
prefill_tok_s
Value
526
Unit
tok/s (pp16k, per-request batch-1)
Model
Llama 3.3 70B (Q4_K)
Methodology
THIRD-PARTY MEASURED (Hardware Corner GPU LLM Benchmarks page, tables dated March 2026). llama.cpp llama-bench, GGUF, single GPU, batch-1. NVIDIA RTX 6000 Ada. Llama 3.3 70B (Q4_K). Number is the 16k-context column of the Prompt Processing table: context depth the run was measured at, not a fixed pp512/tg128. Per-request, NOT aggregate.
Measured by
third_party
Reproducible
yes