Back to search results
Llama-3.1-8B-Instruct · Q4_K_S
NVIDIA H100 80GB HBM380GBllama.cpp v0.22L2 Cross-framework verified
This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (Q4_K_S) on NVIDIA H100 80GB HBM3 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.
314.55 tok/s
Decode
Decode speed
1,364.71 tok/s
Prefill
Prefill speed
2.07 s
TTFT
Time to first token
15.2 GB
VRAM
VRAM usage
L2 Cross-framework verified
1
Measured runs
1
Independent sources
Other
Source platforms
16 days ago
Last verified
Performance
- llama.cpp · IQ1_S Decode 337.53 · Prefill 1795.74 ·
- llama.cpp · Q4_K_S (current)Decode 314.55 · Prefill 1364.71 ·
- Strata · Q5_K_M Decode 362.58 · Prefill 1954.02 ·
Core figures
- Decode (avg)
- 314.55 tok/s
- Prefill (avg)
- 1,364.71 tok/s
- TTFT (avg)
- 2.07 s
- VRAM (avg)
- 15.2 GB
- MTP acceptance rate
- —
- TTFB
- 2.52 GB
- Power draw
- 680.8 W
Configuration
Member-level fields are taken from the most recent run
- Model
- Llama-3.1-8B-Instruct
- Quantization
- Q4_K_S
- Framework
- llama.cpp
- Version
- v0.22
- Context length
- 16384 tokens
- Batch size
- 512
- GPU layers
- 33
- Flash Attention
- Off
Hardware
Nominal and measured figures are shown side by side; whether it runs is the reader's call
- GPU
- NVIDIA H100 80GB HBM3
- Nominal VRAM
- 80 GB
- Measured VRAM (avg)
- 15.2 GB
- OS
- Ubuntu 24.04
- Driver
- 572.83
- CUDA
- 12.8
- Power draw
- 680.8 W
Sources & evidence
1 measured records in total, each traceable to its original source
v0.22 · Ubuntu 24.04 · CUDA 12.8 · 16384 ctx
314.55 tok/s
Decode
1,364.71 tok/s
Prefill
2.07 s
TTFT
15.2 GB
VRAM
—
MTP
680.8 W
Power draw
L2 batch seed 2026-10-19: llama.cpp on hw_id=55 running model_id=68 at Q4_K_S