Arao AI
Back to search results

Llama-3.1-8B-Instruct · IQ1_S

Start a discussion
NVIDIA H100 80GB HBM380GBllama.cpp v0.53L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (IQ1_S) on NVIDIA H100 80GB HBM3 with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

337.53 tok/s

Decode

Decode speed

1,795.74 tok/s

Prefill

Prefill speed

0.24 s

TTFT

Time to first token

9.16 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

16 days ago

Last verified

Performance

  1. llama.cpp · IQ1_S (current)Decode 337.53 · Prefill 1795.74 ·
  2. llama.cpp · Q4_K_S Decode 314.55 · Prefill 1364.71 ·
  3. Strata · Q5_K_M Decode 362.58 · Prefill 1954.02 ·

Core figures

Decode (avg)
337.53 tok/s
Prefill (avg)
1,795.74 tok/s
TTFT (avg)
0.24 s
VRAM (avg)
9.16 GB
MTP acceptance rate
—
TTFB
2.7 GB
Power draw
579 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.1-8B-Instruct
Quantization
IQ1_S
Framework
llama.cpp
Version
v0.53
Context length
16384 tokens
Batch size
256
GPU layers
99
Flash Attention
On

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)
9.16 GB
OS
Ubuntu 24.04
Driver
572.83
CUDA
13.0
Power draw
579 W

Sources & evidence

1 measured records in total, each traceable to its original source

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-19

    v0.53 · Ubuntu 24.04 · CUDA 13.0 · 16384 ctx

    337.53 tok/s

    Decode

    1,795.74 tok/s

    Prefill

    0.24 s

    TTFT

    9.16 GB

    VRAM

    —

    MTP

    579 W

    Power draw

    L2 batch seed 2026-10-19: llama.cpp on hw_id=55 running model_id=68 at IQ1_S