Arao AI
Back to search results

Qwen3-1.7B · IQ4_XS

Start a discussion
Apple M1 MaxStrata v0.50L2 Cross-framework verified

This page aggregates 1 real-world runs of Qwen3-1.7B (IQ4_XS) on Apple M1 Max with Strata, contributed by 1 independent source platforms; metrics are averages of published measurements.

217.54 tok/s

Decode

Decode speed

1,114.27 tok/s

Prefill

Prefill speed

0.85 s

TTFT

Time to first token

4.75 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 · Q4_K_S Decode 157.62 · Prefill 743.87 ·
  2. Strata · IQ4_XS (current)Decode 217.54 · Prefill 1114.27 ·
  3. Strata · Q6_K Decode 152.31 · Prefill 609.9 ·

Core figures

Decode (avg)
217.54 tok/s
Prefill (avg)
1,114.27 tok/s
TTFT (avg)
0.85 s
VRAM (avg)
4.75 GB
MTP acceptance rate
0.65
TTFB
1.74 GB
Power draw
42.9 W

Configuration

Member-level fields are taken from the most recent run

Model
Qwen3-1.7B
Quantization
IQ4_XS
Framework
Strata
Version
v0.50
Context length
4096 tokens
Batch size
1024
GPU layers
12
Flash Attention
Off

Hardware

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

GPU
Apple M1 Max
Measured VRAM (avg)
4.75 GB
OS
macOS 15.0
Driver
Apple M1 Max
CUDA
N/A
Power draw
42.9 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

    v0.50 · macOS 15.0 · CUDA N/A · 4096 ctx

    217.54 tok/s

    Decode

    1,114.27 tok/s

    Prefill

    0.85 s

    TTFT

    4.75 GB

    VRAM

    0.65

    MTP

    42.9 W

    Power draw

    L2 batch seed 2026-10-16: Strata on hw_id=101 running model_id=55 at IQ4_XS