Arao AI
Back to search results

Flint-40B-A2.5B · Q2_K

Start a discussion
Apple M5 Ultra512GBOllama v0.53L2 Cross-framework verified

This page aggregates 1 real-world runs of Flint-40B-A2.5B (Q2_K) on Apple M5 Ultra with Ollama, contributed by 1 independent source platforms; metrics are averages of published measurements.

230.73 tok/s

Decode

Decode speed

1,077.9 tok/s

Prefill

Prefill speed

0.4 s

TTFT

Time to first token

48.07 GB

VRAM

VRAM usage

L2 Cross-framework verified

1

Measured runs

1

Independent sources

Other

Source platforms

14 days ago

Last verified

Performance

  1. llama.cpp · IQ1_S Decode 324.79 · Prefill 1453.99 ·
  2. llama.cpp · Q5_K_S Decode 181.42 · Prefill 831.96 ·
  3. Ollama · Q2_K (current)Decode 230.73 · Prefill 1077.9 ·

Core figures

Decode (avg)
230.73 tok/s
Prefill (avg)
1,077.9 tok/s
TTFT (avg)
0.4 s
VRAM (avg)
48.07 GB
MTP acceptance rate
—
TTFB
1.85 GB
Power draw
127.5 W

Configuration

Member-level fields are taken from the most recent run

Model
Flint-40B-A2.5B
Quantization
Q2_K
Framework
Ollama
Version
v0.53
Context length
16384 tokens
Batch size
256
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
Apple M5 Ultra
Nominal VRAM
512 GB
Measured VRAM (avg)
48.07 GB
OS
macOS 26.1
Power draw
127.5 W

Get started

Original model resources

Base model resources may not match this quantization. Use the verified deployment weights above when available.

Sources & evidence

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

  1. L2 Cross-framework verifiedOtherOriginal link Verified on 2026-10-22

    v0.53 · macOS 26.1 · 16384 ctx

    230.73 tok/s

    Decode

    1,077.9 tok/s

    Prefill

    0.4 s

    TTFT

    48.07 GB

    VRAM

    —

    MTP

    127.5 W

    Power draw

    L2 batch seed 2026-10-22: Ollama on hw_id=29 running model_id=92 at Q2_K