Arao AI
Back to search results

Qwen3.5-9B · IQ2_XS

Start a discussion
NVIDIA Jetson AGX Orin 64GB64GBllama.cpp v0.42L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3.5-9B (IQ2_XS) on NVIDIA Jetson AGX Orin 64GB with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

71.62 tok/s

Decode

Decode speed

230.5 tok/s

Prefill

Prefill speed

0.26 s

TTFT

Time to first token

11.17 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

5 days ago

Last verified

Performance

  1. Ollama · Q3_K_S Decode 55.39 · Prefill 272.09 ·
  2. Strata · Q4_K_M Decode 72.01 · Prefill 281.13 ·
  3. llama.cpp · IQ2_XS (current)Decode 71.62 · Prefill 230.5 ·
  4. llama.cpp · Q4_K_M Decode 47.21 · Prefill 254.51 ·
  5. Strata · Q4_K_S Decode 76.44 · Prefill 335.52 ·
  6. Strata · Q6_K Decode 51.24 · Prefill 260.03 ·

Core figures

Decode (avg)
71.62 tok/s
Prefill (avg)
230.5 tok/s
TTFT (avg)
0.26 s
VRAM (avg)
11.17 GB
MTP acceptance rate
—
TTFB
0.57 GB
Power draw
32.4 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3.5-9B
Quantization
IQ2_XS
Framework
llama.cpp
Version
v0.42
Context length
16384 tokens
Batch size
512
GPU layers
33
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)
11.17 GB
OS
JetPack 6.0
Driver
NVIDIA L4T 36.3
CUDA
12.6
Power draw
32.4 W

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-13

    v0.42 · JetPack 6.0 · CUDA 12.6 · 16384 ctx

    71.62 tok/s

    Decode

    230.5 tok/s

    Prefill

    0.26 s

    TTFT

    11.17 GB

    VRAM

    —

    MTP

    32.4 W

    Power draw

    L2 batch seed 2026-10-13: llama.cpp on hw_id=76 running model_id=49 at IQ2_XS