Arao AI
Back to search results

Qwen3-8B · Q3_K_L

Start a discussion
NVIDIA Jetson AGX Orin 64GB64GBOllama v0.25L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-8B (Q3_K_L) on NVIDIA Jetson AGX Orin 64GB with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

56.02 tok/s

Decode

Decode speed

230.05 tok/s

Prefill

Prefill speed

1.36 s

TTFT

Time to first token

14.2 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. llama.cpp · IQ1_S Decode 80.69 · Prefill 386.89 ·
  2. llama.cpp · IQ2_XS Decode 81.76 · Prefill 342.2 ·
  3. Ollama · FP16 Decode 24.64 · Prefill 90.65 ·
  4. Ollama · Q3_K_L (current)Decode 56.02 · Prefill 230.05 ·
  5. vLLM · Q3_K_S Decode 73.73 · Prefill 262.59 ·
  6. vLLM · Q4_K_S Decode 62 · Prefill 313.33 ·

Core figures

Decode (avg)
56.02 tok/s
Prefill (avg)
230.05 tok/s
TTFT (avg)
1.36 s
VRAM (avg)
14.2 GB
MTP acceptance rate
—
TTFB
0.45 GB
Power draw
40.2 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-8B
Quantization
Q3_K_L
Framework
Ollama
Version
v0.25
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 Jetson AGX Orin 64GB
Nominal VRAM
64 GB
Measured VRAM (avg)
14.2 GB
OS
JetPack 6.0
Driver
NVIDIA L4T 36.3
CUDA
12.6
Power draw
40.2 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-16

    v0.25 · JetPack 6.0 · CUDA 12.6 · 16384 ctx

    56.02 tok/s

    Decode

    230.05 tok/s

    Prefill

    1.36 s

    TTFT

    14.2 GB

    VRAM

    —

    MTP

    40.2 W

    Power draw

    L2 batch seed 2026-10-16: Ollama on hw_id=76 running model_id=53 at Q3_K_L