Arao AI
Back to search results

Llama-3-8B-Instruct · IQ3_S

Start a discussion
NVIDIA RTX 309024GBOllama v0.24L2 Cross-framework verified

This page aggregates 1 real-world runs of Llama-3-8B-Instruct (IQ3_S) on NVIDIA RTX 3090 with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

90.89 tok/s

Decode

Decode speed

298.18 tok/s

Prefill

Prefill speed

1.58 s

TTFT

Time to first token

13.19 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

Today

Last verified

Performance

  1. Ollama · IQ3_S (current)Decode 90.89 · Prefill 298.18 ·
  2. Ollama · Q4_K_M Decode 84.07 · Prefill 419.77 ·
  3. Strata · IQ4_XS Decode 126.45 · Prefill 522.82 ·
  4. Strata · Q3_K_M Decode 136.06 · Prefill 696.33 ·
  5. llama.cpp · Q8_0 Decode 84.9 · Prefill 636.8 ·
  6. llama.cpp · Q5_K_M Decode 123.7 · Prefill 927.8 ·
  7. llama.cpp · Q4_K_M Decode 149 · Prefill 1117.5 ·

Core figures

Decode (avg)
90.89 tok/s
Prefill (avg)
298.18 tok/s
TTFT (avg)
1.58 s
VRAM (avg)
13.19 GB
MTP acceptance rate
—
TTFB
0.73 GB
Power draw
318.7 W

Configuration

Member-level fields are taken from the most recent run

Model
Llama-3-8B-Instruct
Quantization
IQ3_S
Framework
Ollama
Version
v0.24
Context length
8192 tokens
Batch size
512
GPU layers
20
Flash Attention
On

Hardware

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

GPU
NVIDIA RTX 3090
Nominal VRAM
24 GB
Measured VRAM (avg)
13.19 GB
OS
Ubuntu 24.04
Driver
560.94
CUDA
12.6
Power draw
318.7 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

    v0.24 · Ubuntu 24.04 · CUDA 12.6 · 8192 ctx

    90.89 tok/s

    Decode

    298.18 tok/s

    Prefill

    1.58 s

    TTFT

    13.19 GB

    VRAM

    —

    MTP

    318.7 W

    Power draw

    L2 batch seed 2026-10-18: Ollama on hw_id=2 running model_id=69 at IQ3_S