Arao AI
Back to search results

Llama-3.1-8B-Instruct · IQ4_XS

Start a discussion
NVIDIA RTX 509032GBvLLM v0.46L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3.1-8B-Instruct (IQ4_XS) on NVIDIA RTX 5090 with vLLM, contributed by 1 independent source platforms; metrics are averages of published measurements.

310.8 tok/s

Decode

Decode speed

1,667.93 tok/s

Prefill

Prefill speed

2.16 s

TTFT

Time to first token

14.53 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. vLLM · IQ4_XS (current)Decode 310.8 · Prefill 1667.93 ·
  2. Ollama · IQ1_S Decode 316.88 · Prefill 1203.93 ·
  3. llama.cpp · Q8_0 Decode 162.5 · Prefill 1218.8 ·
  4. llama.cpp · Q5_K_M Decode 236.9 · Prefill 1776.8 ·
  5. llama.cpp · Q4_K_M Decode 285.3 · Prefill 2139.8 ·

Core figures

Decode (avg)
310.8 tok/s
Prefill (avg)
1,667.93 tok/s
TTFT (avg)
2.16 s
VRAM (avg)
14.53 GB
MTP acceptance rate
—
TTFB
2.49 GB
Power draw
457 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3.1-8B-Instruct
Quantization
IQ4_XS
Framework
vLLM
Version
v0.46
Context length
32768 tokens
Batch size
32
GPU layers
999
Flash Attention
On

Hardware

Nominal and measured figures are shown side by side; whether it runs is the reader's call

GPU
NVIDIA RTX 5090
Nominal VRAM
32 GB
Measured VRAM (avg)
14.53 GB
OS
Windows 11 24H2
Driver
565.57
CUDA
12.4
Power draw
457 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-18

    v0.46 · Windows 11 24H2 · CUDA 12.4 · 32768 ctx

    310.8 tok/s

    Decode

    1,667.93 tok/s

    Prefill

    2.16 s

    TTFT

    14.53 GB

    VRAM

    —

    MTP

    457 W

    Power draw

    L2 batch seed 2026-10-18: vLLM on hw_id=13 running model_id=68 at IQ4_XS