Arao AI
Back to search results

Qwen2.5-0.5B-Instruct · IQ1_S

Start a discussion
Apple M5 Max128GBStrata v0.38L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen2.5-0.5B-Instruct (IQ1_S) on Apple M5 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

343.91 tok/s

Decode

Decode speed

1,513.7 tok/s

Prefill

Prefill speed

2.2 s

TTFT

Time to first token

2.54 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

2 days ago

Last verified

Performance

  1. vLLM · IQ2_XS Decode 291.82 · Prefill 1370.02 ·
  2. Strata · IQ1_S (current)Decode 343.91 · Prefill 1513.7 ·
  3. Strata · Q5_K_S Decode 305.21 · Prefill 1469.7 ·
  4. llama.cpp · IQ2_M Decode 298.63 · Prefill 1141.07 ·
  5. llama.cpp · Q8_0 Decode 366.43 · Prefill 1208.56 ·
  6. vLLM · Q2_K Decode 318.68 · Prefill 1287.78 ·

Core figures

Decode (avg)
343.91 tok/s
Prefill (avg)
1,513.7 tok/s
TTFT (avg)
2.2 s
VRAM (avg)
2.54 GB
MTP acceptance rate
0.48
TTFB
2.75 GB
Power draw
270.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen2.5-0.5B-Instruct
Quantization
IQ1_S
Framework
Strata
Version
v0.38
Context length
8192 tokens
Batch size
512
GPU layers
8
Flash Attention
On

Hardware

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

GPU
Apple M5 Max
Nominal VRAM
128 GB
Measured VRAM (avg)
2.54 GB
OS
Ubuntu 24.04
Driver
560.94
CUDA
12.4
Power draw
270.5 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

    v0.38 · Ubuntu 24.04 · CUDA 12.4 · 8192 ctx

    343.91 tok/s

    Decode

    1,513.7 tok/s

    Prefill

    2.2 s

    TTFT

    2.54 GB

    VRAM

    0.48

    MTP

    270.5 W

    Power draw

    L2 batch seed 2026-10-16: Strata on hw_id=28 running model_id=59 at IQ1_S