Arao AI
Back to search results

Qwen2.5-0.5B-Instruct · Q3_K_S

Start a discussion
NVIDIA Jetson AGX Orin 64GB64GBllama.cpp v0.21L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-0.5B-Instruct (Q3_K_S) on NVIDIA Jetson AGX Orin 64GB with llama.cpp, contributed by 1 independent source platforms; metrics are averages of published measurements.

199.9 tok/s

Decode

Decode speed

998.9 tok/s

Prefill

Prefill speed

0.95 s

TTFT

Time to first token

2.78 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 243.52 · Prefill 900.17 ·
  2. llama.cpp · Q3_K_S (current)Decode 199.9 · Prefill 998.9 ·
  3. Strata · IQ3_XS Decode 310.9 · Prefill 1483.77 ·
  4. Strata · IQ2_M Decode 337.39 · Prefill 1690.81 ·

Core figures

Decode (avg)
199.9 tok/s
Prefill (avg)
998.9 tok/s
TTFT (avg)
0.95 s
VRAM (avg)
2.78 GB
MTP acceptance rate
—
TTFB
1.6 GB
Power draw
51.6 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-0.5B-Instruct
Quantization
Q3_K_S
Framework
llama.cpp
Version
v0.21
Context length
32768 tokens
Batch size
256
GPU layers
28
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)
2.78 GB
OS
JetPack 6.0
Driver
NVIDIA L4T 36.3
CUDA
12.6
Power draw
51.6 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

    v0.21 · JetPack 6.0 · CUDA 12.6 · 32768 ctx

    199.9 tok/s

    Decode

    998.9 tok/s

    Prefill

    0.95 s

    TTFT

    2.78 GB

    VRAM

    —

    MTP

    51.6 W

    Power draw

    L2 batch seed 2026-10-16: llama.cpp on hw_id=76 running model_id=59 at Q3_K_S