Hardware Corner Mar 2026: RTX 3080 Ti Qwen3 14B Q4_K decode @ 16k ctx
52.3 tok/s (decode, per-request, batch 1, KV depth 16k)
Fields
id
benchmarks/hc26-rtx3080ti-qwen3-14b-q4k-tg-16kMetric
decode_tok_s
Value
52.3
Unit
tok/s (decode, per-request, batch 1, KV depth 16k)
Model
Qwen3 14B Q4_K
Methodology
THIRD-PARTY LAB MEASURED (Hardware Corner per-GPU LLM benchmark page, Updated Benchmarks: March 2026). llama.cpp, single GPU, weights fully in VRAM. Context length is the KV depth the table column names. Token Generation column = batch-1 per-request decode tok/s, NOT aggregate across concurrent requests. Prompt Processing column = prefill tok/s at that context depth. Engine build hash, CUDA version and exact prompt/output token counts beyond the context-depth bucket are not published. Where the page header quotes a single '14B @ 4k' figure that disagrees with the per-model table, the per-cell table value is the one recorded.
Measured by
third_party
Reproducible
yes
Thermal state
unknown
notes
Single RTX 3080 Ti 12GB GDDR6X, 912 GB/s as stated on the page. Ampere, PCIe gen4. 12 GB caps the card at 14B Q4_K / 16k despite the bandwidth. format_id is the nearest standard GGUF; the page labels the quant Q4_K. Key-insights line quotes 52.3 t/s at 16k, matching this cell. Header '52.2 @ 4k' is this number rounded and mislabeled.
Notes
Single RTX 3080 Ti 12GB GDDR6X, 912 GB/s as stated on the page. Ampere, PCIe gen4. 12 GB caps the card at 14B Q4_K / 16k despite the bandwidth. format_id is the nearest standard GGUF; the page labels the quant Q4_K. Key-insights line quotes 52.3 t/s at 16k, matching this cell. Header '52.2 @ 4k' is this number rounded and mislabeled.